From DoNotReply at t10.org Sat Jul 4 00:02:22 2020 From: DoNotReply at t10.org (DoNotReply at t10.org) Date: Sat, 04 Jul 2020 00:02:22 -0700 Subject: [T10] Recent T10 documents uploaded since 2020/06/27 Message-ID: <01d651d1$Blat.v2.2.2$12672e6f@wdc.com> Proposals --------- SPC-6: Updated Command Duration Limits Processing Optimizations (by: Ralph Weber) T10/20-056r1 Uploaded: 2020/06/28 137330 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-056r1.pdf UAS-3: ERDY Clarifications (by: Ralph Weber) T10/20-066r0 Uploaded: 2020/06/29 73049 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-066r0.pdf ZAC-2 Add rules for depopulation interaction with zoned block devices (by: Gerald Houlder) T10/20-069r0 Uploaded: 2020/06/30 9326 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-069r0.pdf SBC-5: Device Server Sequestered Commands (by: Ralph Weber) T10/20-072r0 Uploaded: 2020/07/03 115536 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-072r0.pdf Working Drafts -------------- (Report generated on 2020/07/04 at 00:02:04) From Ralph.Weber at wdc.com Tue Jul 7 10:21:58 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Tue, 7 Jul 2020 17:21:58 +0000 Subject: [T10] Network access changes to t10.org Message-ID: After lengthy internal negotiations among numerous WD stakeholders, this coming weekend has been selected for a reconfiguration of the networking interfaces to a large collection of virtual machines housed behind the WD Firewall. The webservers for t10.org and t13.org will be affected by this reconfiguration. The most dramatic effects are possible on Saturday (11 July) in the Pacific Cost Time between mid-morning and early afternoon. During this time, there is a possibility that the webservers will be so inaccessible that a command line Ping to either of them will timeout (e.g., C:> ping t13.org will be unsuccessful). Both before and after the dark hours of the inaccessible interval, changes are possible that will confuse web browsers to an extent that makes the websites "look" inaccessible. These conditions arise from the fact that (in addition to all the other cached goodies) web browsers cache translations between URLs and their underlying IP addresses. The resulting web browsing behavior can be bewildering, to say the least. Complaints such as, "It works when I disable the VPN but not when the VPN is on" suggest the extent of the possible turmoil. For this reason, I have been asked to distribute the following URL to provide guidance regarding how to clear the just-right part of a browser's cache: http://www.wikihow.com/Clear-Your-Browser's-Cache For T10, the proximity to the July Plenary Week is unfortunate, but quite unavoidable. "Plan ahead accordingly", lame though it may be, is the best advice available. Suffice to say that voices which carry much more weight than our intrepid band of standards developers had lots to say about the choice of this weekend. All the best, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From Scott.Shuey at luxshare-ict.com Wed Jul 8 06:44:22 2020 From: Scott.Shuey at luxshare-ict.com (Scott Shuey(LUXSHARE-ICT)) Date: Wed, 8 Jul 2020 13:44:22 +0000 Subject: [T10] Network access changes to t10.org In-Reply-To: References: Message-ID: <3f07fd7ad06747be9f9c800d0c4ab5c5@TP-MBS-01.luxshare.com.cn> Ralph, Where's the invite and call-in information for the Plenary? I don't see anything and it's not on my calendar as an invite. Did one go out? Regards, From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Tuesday, July 7, 2020 1:22 PM To: t10 at t10.org Subject: [T10] Network access changes to t10.org After lengthy internal negotiations among numerous WD stakeholders, this coming weekend has been selected for a reconfiguration of the networking interfaces to a large collection of virtual machines housed behind the WD Firewall. The webservers for t10.org and t13.org will be affected by this reconfiguration. The most dramatic effects are possible on Saturday (11 July) in the Pacific Cost Time between mid-morning and early afternoon. During this time, there is a possibility that the webservers will be so inaccessible that a command line Ping to either of them will timeout (e.g., C:> ping t13.org will be unsuccessful). Both before and after the dark hours of the inaccessible interval, changes are possible that will confuse web browsers to an extent that makes the websites "look" inaccessible. These conditions arise from the fact that (in addition to all the other cached goodies) web browsers cache translations between URLs and their underlying IP addresses. The resulting web browsing behavior can be bewildering, to say the least. Complaints such as, "It works when I disable the VPN but not when the VPN is on" suggest the extent of the possible turmoil. For this reason, I have been asked to distribute the following URL to provide guidance regarding how to clear the just-right part of a browser's cache: http://www.wikihow.com/Clear-Your-Browser's-Cache For T10, the proximity to the July Plenary Week is unfortunate, but quite unavoidable. "Plan ahead accordingly", lame though it may be, is the best advice available. Suffice to say that voices which carry much more weight than our intrepid band of standards developers had lots to say about the choice of this weekend. All the best, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From joe at lingua-data.com Wed Jul 8 09:14:07 2020 From: joe at lingua-data.com (Joe Breher) Date: Wed, 8 Jul 2020 10:14:07 -0600 Subject: [T10] Cause Celebre Message-ID: <7F667A72-2638-4073-B364-B79AD1417232@lingua-data.com> This just in: Data storage protocol wonks vindicated as Merriam-Webster adds ?embiggen? to the dictionary: https://www.mentalfloss.com/article/534500/embiggen-the-simpsons-word-merriam-webster-dictionary Joe Breher (478) 2-Breher (478) 227-3437 -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill.martin at samsung.com Wed Jul 8 16:28:59 2020 From: bill.martin at samsung.com (Bill Martin) Date: Wed, 8 Jul 2020 23:28:59 +0000 Subject: [T10] Bill Martin (Samsung)'s Zoom Meeting References: Message-ID: [https://d24cgw3uvb9a9h.cloudfront.net/static/94014/image/new/ZoomLogo_110_25.png] Hi there, Bill Martin (Samsung) is inviting you to a scheduled Zoom meeting. Join Zoom Meeting Phone one-tap: US: +12133388477,,96852292749#,,,,0#,,58674957# or +16692192599,,96852292749#,,,,0#,,58674957# Meeting URL: https://samsung-dsa.zoom.us/j/96852292749?pwd=VUxPdGdtNjIwam50Z0RGalNnTlpoZz09 Meeting ID: 968 5229 2749 Password: 9j?KKzF% Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 213 338 8477 or +1 669 219 2599 or +1 669 900 9128 or +1 346 248 7799 or +1 602 753 0140 or +1 720 928 9299 or +1 971 247 1195 or +1 253 215 8782 or +1 267 831 0333 or +1 301 715 8592 or +1 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 or +1 646 518 9805 or +1 646 558 8656 or +1 651 372 8299 or +1 786 635 1003 Meeting ID: 968 5229 2749 Password: 58674957 International numbers Join from an H.323/SIP room system H.323: 162.255.37.11 (US West) 162.255.36.11 (US East) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 968 5229 2749 Password: 58674957 SIP: 96852292749 at zoomcrc.com Password: 58674957 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3290 bytes Desc: not available URL: From bill.martin at samsung.com Wed Jul 8 16:30:08 2020 From: bill.martin at samsung.com (Bill Martin) Date: Wed, 8 Jul 2020 23:30:08 +0000 Subject: [T10] T10 Plenary meeting References: Message-ID: <104b53dc4101406d990573488dfa1699@samsung.com> [https://d24cgw3uvb9a9h.cloudfront.net/static/94014/image/new/ZoomLogo_110_25.png] Hi there, Bill Martin (Samsung) is inviting you to a scheduled Zoom meeting. Join Zoom Meeting Phone one-tap: US: +16699009128,,93761004082#,,,,0#,,99699312# or +12133388477,,93761004082#,,,,0#,,99699312# Meeting URL: https://samsung-dsa.zoom.us/j/93761004082?pwd=RnZRS3JMbUdURDZVc0Z0V0dtTHhaZz09 Meeting ID: 937 6100 4082 Password: 6k+6^B6K Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 669 900 9128 or +1 213 338 8477 or +1 669 219 2599 or +1 720 928 9299 or +1 971 247 1195 or +1 253 215 8782 or +1 346 248 7799 or +1 602 753 0140 or +1 646 518 9805 or +1 646 558 8656 or +1 651 372 8299 or +1 786 635 1003 or +1 267 831 0333 or +1 301 715 8592 or +1 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 Meeting ID: 937 6100 4082 Password: 99699312 International numbers Join from an H.323/SIP room system H.323: 162.255.37.11 (US West) 162.255.36.11 (US East) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 937 6100 4082 Password: 99699312 SIP: 93761004082 at zoomcrc.com Password: 99699312 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3236 bytes Desc: not available URL: From bill.martin at samsung.com Wed Jul 8 16:31:08 2020 From: bill.martin at samsung.com (Bill Martin) Date: Wed, 8 Jul 2020 23:31:08 +0000 Subject: [T10] T10 CAP working group meeting References: Message-ID: [https://d24cgw3uvb9a9h.cloudfront.net/static/94014/image/new/ZoomLogo_110_25.png] Hi there, Bill Martin (Samsung) is inviting you to a scheduled Zoom meeting. Join Zoom Meeting Phone one-tap: US: +12133388477,,96852292749#,,,,0#,,58674957# or +16692192599,,96852292749#,,,,0#,,58674957# Meeting URL: https://samsung-dsa.zoom.us/j/96852292749?pwd=VUxPdGdtNjIwam50Z0RGalNnTlpoZz09 Meeting ID: 968 5229 2749 Password: 9j?KKzF% Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 213 338 8477 or +1 669 219 2599 or +1 669 900 9128 or +1 346 248 7799 or +1 602 753 0140 or +1 720 928 9299 or +1 971 247 1195 or +1 253 215 8782 or +1 267 831 0333 or +1 301 715 8592 or +1 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 or +1 646 518 9805 or +1 646 558 8656 or +1 651 372 8299 or +1 786 635 1003 Meeting ID: 968 5229 2749 Password: 58674957 International numbers Join from an H.323/SIP room system H.323: 162.255.37.11 (US West) 162.255.36.11 (US East) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 968 5229 2749 Password: 58674957 SIP: 96852292749 at zoomcrc.com Password: 58674957 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3283 bytes Desc: not available URL: From curtis.stevens at seagate.com Wed Jul 8 16:31:26 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Wed, 8 Jul 2020 23:31:26 +0000 Subject: [T10] SPC-6r2 is now posted Message-ID: Also posted is the issues list, 19-038r2. Enjoy! --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Fri Jul 10 13:38:15 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Fri, 10 Jul 2020 20:38:15 +0000 Subject: [T10] SAT-5 is now posted Message-ID: SAT-5 Rev 4 is now posted along with the issues list 19-033r4. Enjoy! --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Fri Jul 10 14:00:14 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Fri, 10 Jul 2020 21:00:14 +0000 Subject: [T10] {Postponed} t10.org will not be affected by a network reconfiguration until at least August Message-ID: The network reconfiguration planned for 11 July has been postponed until at least August. The previously predicted possibility of service outages no longer is expected. All the best, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill.martin at samsung.com Fri Jul 10 14:18:20 2020 From: bill.martin at samsung.com (Bill Martin) Date: Fri, 10 Jul 2020 21:18:20 +0000 Subject: [T10] Dates for 2021 References: Message-ID: <4d5e3d84e6b840b1b159fa8ffcfaa8be@samsung.com> I have put proposed dates for the 2021 meetings on the T10 meeting page. Please look at these and come prepared to discuss them at the T10 CAP working group next week with final approval at the plenary meeting. Bill Martin Chair INCITS T10 Co-Chair SNIA Technical Council SSD I/O Standards NVMe Board of Directors Samsung Semiconductor, Inc. Cell (408) 499-1839 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Kevin.Marks at dell.com Fri Jul 10 15:14:35 2020 From: Kevin.Marks at dell.com (Kevin.Marks at dell.com) Date: Fri, 10 Jul 2020 22:14:35 +0000 Subject: [T10] CAP Schedule Message-ID: Does anyone have any specific time requests for their topics? Building the schedule this weekend. Kevin Kevin Marks Technologist, Distinguished Engineer Dell EMC | Server & Infrastructure Systems Office of the CTO Work +1-512-723-9173 Mobile +1-512-963-8178 Email Kevin.Marks at Dell.com Dell EMC, One Dell Way, MS PS2-47 Round Rock, TX 78682 USA -------------- next part -------------- An HTML attachment was scrubbed... URL: From DoNotReply at t10.org Sat Jul 11 00:02:23 2020 From: DoNotReply at t10.org (DoNotReply at t10.org) Date: Sat, 11 Jul 2020 00:02:23 -0700 Subject: [T10] Recent T10 documents uploaded since 2020/07/04 Message-ID: <01d65751$Blat.v2.2.2$3b652ed0@wdc.com> Proposals --------- SAT-5: Issues List (by: Curtis E. Stevens) T10/19-033r4 Uploaded: 2020/07/10 85849 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=19-033r4.pdf SPC-6 Issues List (by: Curtis Stevens) T10/19-038r2 Uploaded: 2020/07/08 70842 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=19-038r2.pdf SBC-5: Performance Managed Power Savings (by: Ralph Weber) T10/20-073r0 Uploaded: 2020/07/06 103041 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-073r0.pdf SBC-5: Performance Managed Power Savings (by: Ralph Weber) T10/20-073r1 Uploaded: 2020/07/10 99178 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-073r1.pdf Working Drafts -------------- SCSI / ATA Translation - 5 (Editor: Curtis Stevens) Rev: 04 Uploaded: 2020/07/10 2479843 bytes https://www.t10.org/cgi-bin/ac.pl?t=f&f=sat5r04.pdf SCSI Primary Commands - 6 (SPC-6) (Editor: Curtis Stevens) Rev: 02 Uploaded: 2020/07/08 6044056 bytes https://www.t10.org/cgi-bin/ac.pl?t=f&f=spc6r02.pdf (Report generated on 2020/07/11 at 00:02:07) From Kevin.Marks at dell.com Sun Jul 12 15:37:19 2020 From: Kevin.Marks at dell.com (Kevin.Marks at dell.com) Date: Sun, 12 Jul 2020 22:37:19 +0000 Subject: [T10] CAP Schedule In-Reply-To: References: Message-ID: <22fd7255944f4b81beb9b56bede881fb@ausx13mps333.AMER.DELL.COM> Initial CAP Schedule From: Marks, Kevin Sent: Friday, July 10, 2020 5:15 PM To: 't10 at t10.org' Subject: CAP Schedule Does anyone have any specific time requests for their topics? Building the schedule this weekend. Kevin Kevin Marks Technologist, Distinguished Engineer Dell EMC | Server & Infrastructure Systems Office of the CTO Work +1-512-723-9173 Mobile +1-512-963-8178 Email Kevin.Marks at Dell.com Dell EMC, One Dell Way, MS PS2-47 Round Rock, TX 78682 USA -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: July Cap Schedule.pdf Type: application/pdf Size: 101888 bytes Desc: July Cap Schedule.pdf URL: From gerry.houlder at seagate.com Mon Jul 13 08:41:14 2020 From: gerry.houlder at seagate.com (Gerry Houlder) Date: Mon, 13 Jul 2020 15:41:14 +0000 Subject: [T10] CAP schedule: 19-066r6 will be withdrawn Message-ID: 19-066r6 will be withdrawn. The CAP discussion schedule can be redone to allow for more time for other proposals, longer breaks, or both. -------------- next part -------------- An HTML attachment was scrubbed... URL: From lohmeyerjohn at gmail.com Mon Jul 13 10:03:53 2020 From: lohmeyerjohn at gmail.com (John Lohmeyer) Date: Mon, 13 Jul 2020 11:03:53 -0600 Subject: [T10] Please ignore this test message Message-ID: <41eb3f77-fc26-72ed-5756-90b91071a94a@gmail.com> This is just further verification that the T10 reflector is still working... From curtis.ballard at hpe.com Mon Jul 13 15:18:06 2020 From: curtis.ballard at hpe.com (Ballard, Curtis C (HPE Storage)) Date: Mon, 13 Jul 2020 22:18:06 +0000 Subject: [T10] T10 20-041r2 is withdrawn Message-ID: I was just looking at the T10 CAP scheduled and noticed that both 20-041r2 and 20-061r0 are listed on the agenda. I meant to request that 20-041r2 be withdrawn following the May CAP meetings as the group decided that the model proposed in 20-061r0 was preferred and both addressed the same problem space. Curtis Ballard Hewlett Packard Enterprise -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Thu Jul 16 16:26:41 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 16 Jul 2020 23:26:41 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Message-ID: Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From Frederick.Knight at netapp.com Thu Jul 16 19:09:51 2020 From: Frederick.Knight at netapp.com (Knight, Frederick) Date: Fri, 17 Jul 2020 02:09:51 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: To me, RMB is only about media, and has nothing to do with removing the device server or task manager (or anything else that is not media). Removal of the device server or task manager (and the associated LUR or I_T Nexus Loss) has nothing to do with the RMB bit. This kind of description seems to mix those things together. You also left out the case where RMB=0 indicates that the medium is not removable without resulting in a LOGICAL UNIT NOT SUPPORTED (you'll likely get this error long before you get the POR (29/00) UA notice). There are probably others. So it just seems problematic to try and build a list of all the errors that might result if/when an RMB=0 device is removed (because removing the media of an RMB=0 device means you're removing the whole device). Fred From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 7:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition NetApp Security WARNING: This is an external email. Do not click links or open attachments unless you recognize the sender and know the content is safe. Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From Kevin.Marks at dell.com Thu Jul 16 19:16:43 2020 From: Kevin.Marks at dell.com (Kevin.Marks at dell.com) Date: Fri, 17 Jul 2020 02:16:43 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: <0726a36eb0a74d7abe41c2e1561ad9f9@ausx13mps333.AMER.DELL.COM> I concur with Fred. Kevin From: t10-bounces at t10.org On Behalf Of Knight, Frederick Sent: Thursday, July 16, 2020 9:10 PM To: Ralph Weber; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition [EXTERNAL EMAIL] To me, RMB is only about media, and has nothing to do with removing the device server or task manager (or anything else that is not media). Removal of the device server or task manager (and the associated LUR or I_T Nexus Loss) has nothing to do with the RMB bit. This kind of description seems to mix those things together. You also left out the case where RMB=0 indicates that the medium is not removable without resulting in a LOGICAL UNIT NOT SUPPORTED (you'll likely get this error long before you get the POR (29/00) UA notice). There are probably others. So it just seems problematic to try and build a list of all the errors that might result if/when an RMB=0 device is removed (because removing the media of an RMB=0 device means you're removing the whole device). Fred From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 7:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition NetApp Security WARNING: This is an external email. Do not click links or open attachments unless you recognize the sender and know the content is safe. Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.ballard at hpe.com Fri Jul 17 07:53:46 2020 From: curtis.ballard at hpe.com (Ballard, Curtis C (HPE Storage)) Date: Fri, 17 Jul 2020 14:53:46 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From Paul.Suhler at kioxia.com Fri Jul 17 09:14:49 2020 From: Paul.Suhler at kioxia.com (Paul Suhler) Date: Fri, 17 Jul 2020 16:14:49 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From: t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From John.Geldman at kioxia.com Fri Jul 17 09:42:03 2020 From: John.Geldman at kioxia.com (John Geldman) Date: Fri, 17 Jul 2020 16:42:03 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: <84ac78f041594ffd8e04b9aa0c884048@kioxia.com> I concur. At the time, the high-falutin' concept (without resulting in a Logical Unit Reset condition (see SAM-6)) was not close to being in our minds. The concern was reporting that the floppy, CD, or ZIP cartridge may not have been in its slot. John From: t10-bounces at t10.org On Behalf Of Paul Suhler Sent: Friday, July 17, 2020 9:15 AM To: Ballard, Curtis C (HPE Storage) ; Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From: t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From kdbutt at us.ibm.com Fri Jul 17 09:40:22 2020 From: kdbutt at us.ibm.com (Kevin D Butt) Date: Fri, 17 Jul 2020 09:40:22 -0700 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From: t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://urldefense.proofpoint.com/v2/url?u=https-3A__www.t10.org_mailman_listinfo_t10&d=DwICAg&c=jf_iaSHvJObTbx-siA1ZOg&r=9hRV7-xktvhH6SmVBcFXOfc3l09GaQd0rqZQQg8BGkY&m=hRMzkCOP3LQXCstClvfayTnkNYyehv2AMVrc2e-stRk&s=fgRRY0l_c-D228Qyz7mYvOp8l6BLXOk_ukQ8xnPmofQ&e= -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Fri Jul 17 09:59:12 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Fri, 17 Jul 2020 16:59:12 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: , Message-ID: We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Kevin D Butt Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org ; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Mike.Webster at wdc.com Fri Jul 17 10:30:01 2020 From: Mike.Webster at wdc.com (Michael Webster) Date: Fri, 17 Jul 2020 17:30:01 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: Message-ID: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: on behalf of Curtis Stevens Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt , Paul Suhler Cc: "t10 at t10.org" , "t10-bounces at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Kevin D Butt Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org ; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From DoNotReply at t10.org Sat Jul 18 00:02:22 2020 From: DoNotReply at t10.org (DoNotReply at t10.org) Date: Sat, 18 Jul 2020 00:02:22 -0700 Subject: [T10] Recent T10 documents uploaded since 2020/07/11 Message-ID: <01d65cd1$Blat.v2.2.2$642ef532@wdc.com> Proposals --------- ZBC-2: Issues List (by: Ralph Weber) T10/19-008r9 Uploaded: 2020/07/14 65873 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=19-008r9.pdf T13 Liaison Report to T10 (by: Kevin Marks) T10/19-017r9 Uploaded: 2020/07/15 109236 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=19-017r9.pdf SPC-6: Updated Command Duration Limits Processing Optimizations (by: Ralph Weber) T10/20-056r2 Uploaded: 2020/07/14 132398 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-056r2.pdf T11 Liaison Report April 2020 (by: Steve Wilson) T10/20-060r1 Uploaded: 2020/07/15 119358 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-060r1.pdf T11 Liaison Report June 2020 (by: Steve Wilson) T10/20-060r2 Uploaded: 2020/07/15 119988 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-060r2.pdf STA Liaison Report to T10 July 2020 (by: Cameron Brett) T10/20-067r1 Uploaded: 2020/07/16 204321 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-067r1.pdf STA Liaison Report to T10 July 2020 (by: Cameron Brett) T10/20-067r1 Uploaded: 2020/07/16 56857 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-067r1.pptx ZBC-2 Add rules for depopulation interaction with zoned block devices (by: Gerald Houlder) T10/20-069r1 Uploaded: 2020/07/14 9384 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-069r1.pdf SSC-5 IBM missing volume stop writes log parameters RFC comment resolution (by: Kevin Butt) T10/20-070r1 Uploaded: 2020/07/14 28478 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-070r1.pdf SAM-6: Fixing lost UA fix (by: William Martin) T10/20-076r0 Uploaded: 2020/07/14 602542 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-076r0.pdf SAM-6: Fixing lost UA fix (by: William Martin) T10/20-076r1 Uploaded: 2020/07/14 602543 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-076r1.pdf SAT-5 WWN updates (by: Brad Besmer) T10/20-077r0 Uploaded: 2020/07/14 91368 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-077r0.pdf SAT-5 WWN updates (by: Brad Besmer) T10/20-077r1 Uploaded: 2020/07/14 91810 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-077r1.pdf SAT-5: ATA SMART EXECUTE OFF-LINE obsolete (by: Brad Besmer) T10/20-078r0 Uploaded: 2020/07/15 144658 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-078r0.pdf SAT-5 Cleanup NCQ Errors (by: Brad Besmer) T10/20-079r0 Uploaded: 2020/07/15 98085 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-079r0.pdf SAT-5 Cleanup Start-Stop Cycle Counter log page (by: Brad Besmer) T10/20-080r0 Uploaded: 2020/07/15 83826 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-080r0.pdf SNIA Liaison Report - July 2020 (by: Frederick Knight) T10/20-081r0 Uploaded: 2020/07/16 763588 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-081r0.pdf Working Drafts -------------- Zoned Block Commands - 2 (ZBC-2) (Editor: Ralph O. Weber) Rev: 06 Uploaded: 2020/07/14 1131022 bytes https://www.t10.org/cgi-bin/ac.pl?t=f&f=zbc2r06.pdf (Report generated on 2020/07/18 at 00:01:59) From Ralph.Weber at wdc.com Tue Jul 21 10:40:51 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Tue, 21 Jul 2020 17:40:51 +0000 Subject: [T10] Network access changes to t10.org Message-ID: This coming Saturday has been selected for a renewed effort to reconfigure the networking interfaces to a large collection of virtual machines housed behind the WD Firewall. The webservers for t10.org and t13.org will be affected by this reconfiguration. Since only ad hoc meetings are scheduled for the next few weeks, the new schedule seems like an improvement for T10 and T13. The most dramatic effects are possible on Saturday (25 July) in the Pacific Cost Time between mid-morning and early afternoon. During this time, there is a possibility that the webservers will be so inaccessible that a command line Ping to either of them will timeout (e.g., C:> ping t13.org will timeout). Both before and after the dark hours of inaccessibility, changes are possible that will confuse web browsers to an extent that makes the websites "look" inaccessible. These conditions arise from the fact that (in addition to all the other cached goodies) web browsers cache translations between URLs and their underlying IP addresses. The resulting web browsing behavior can be bewildering, to say the least. Complaints such as, "It works when I disable the VPN but not when the VPN is on" suggest the extent of the possible turmoil. For this reason, I have been asked to distribute the following URL to provide guidance regarding how to clear the just-right part of a browser's cache: http://www.wikihow.com/Clear-Your-Browser's-Cache All the best, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From Mike.Webster at wdc.com Tue Jul 21 17:56:55 2020 From: Mike.Webster at wdc.com (Michael Webster) Date: Wed, 22 Jul 2020 00:56:55 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> Message-ID: <53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: on behalf of Mike Webster Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: on behalf of Curtis Stevens Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt , Paul Suhler Cc: "t10 at t10.org" , "t10-bounces at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Kevin D Butt Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org ; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Wed Jul 22 11:34:40 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Wed, 22 Jul 2020 18:34:40 +0000 Subject: [T10] UAS-3 Synch With SAM-6 Message-ID: My home-office review of 20-051r1 "UAS-3: Align transport protocol services with SAM-6" produced more discomfiture than CAP could have inflicted in ten hours of discussion. Before the dust settled, every page in the SAM transport functions section had been reviewed no fewer than two times. The CAP guidance last week was excellent, especially the "derived from" wording for the logical unit numbers in I_T_L nexus values, but ... The extent of the changes in 20-051r2 has emphasized the need for an additional round of review. Therefore, a 1-hour WebEx is being hosted by Western Digital on Thursday, 6 August starting at 2 pm EDT. This 20-051r2 review will not run past an hour because, if it does, I'm going back to the drawing board. Having been announced two weeks in advance on the T10 Reflector, the WebEx qualifies as an impromptu T10 Ad Hoc and in the absence of complaints from the T10 officers, it will be so documented on the T10 website. All the best, .Ralph -- Do not delete or change any of the following text. -- Options to Join this Meeting >From any computer or mobile device, Click to Join WebEx. >From any video conference unit, enter "54 169 064 7742" from remote or touch panel. Meeting password: NKc2257362926 (65222573 from phones) ________________________________ Toll and Toll free numbers are billed at a higher per minute rate. Follow the below instructions for cost effective conferencing. 1. Click link above to join WebEx meeting. 2. Once you are in the meeting, click the Phone icon and select Call Using Computer. Or on the WebEx Mobile App, select Use Internet for Audio. 3. If your microphone, camera or screen sharing doesn't work Click Here 4. Only use Call me If using a WD Cisco desk phone, a WD video conference device, or Jabber soft phone app. 5. Questions? Refer to WebEx Calling FAQ's If you are unable to use 'Call Using Computer' or 'Call Me' features, use the following dial in numbers Join by phone +1-408-717-7733 USA Toll Meeting Number: 169 064 7742 Global call-in numbers -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 4166 bytes Desc: not available URL: From bill.martin at samsung.com Fri Jul 24 09:45:39 2020 From: bill.martin at samsung.com (Bill Martin) Date: Fri, 24 Jul 2020 16:45:39 +0000 Subject: [T10] T10 CAP working group References: Message-ID: [https://d24cgw3uvb9a9h.cloudfront.net/static/94026/image/new/ZoomLogo_110_25.png] Hi there, Bill Martin (Samsung) is inviting you to a scheduled Zoom meeting. Join Zoom Meeting Phone one-tap: US: +16699009128,,99403434298#,,,,,,0#,,49645580# or +12133388477,,99403434298#,,,,,,0#,,49645580# Meeting URL: https://samsung-dsa.zoom.us/j/99403434298?pwd=eTdYeC9SQWY0Sk1La2M4Z0hJSHpRZz09 Meeting ID: 994 0343 4298 Passcode: 4i.wejny Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 669 900 9128 or +1 213 338 8477 or +1 669 219 2599 or +1 602 753 0140 or +1 720 928 9299 or +1 971 247 1195 or +1 253 215 8782 or +1 346 248 7799 or +1 786 635 1003 or +1 267 831 0333 or +1 301 715 8592 or +1 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 or +1 646 518 9805 or +1 646 558 8656 or +1 651 372 8299 Meeting ID: 994 0343 4298 Passcode: 49645580 International numbers Join from an H.323/SIP room system H.323: 162.255.37.11 (US West) 162.255.36.11 (US East) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 994 0343 4298 Passcode: 49645580 SIP: 99403434298 at zoomcrc.com Passcode: 49645580 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3302 bytes Desc: not available URL: From bill.martin at samsung.com Fri Jul 24 09:48:21 2020 From: bill.martin at samsung.com (Bill Martin) Date: Fri, 24 Jul 2020 16:48:21 +0000 Subject: [T10] T10 CAP working group References: Message-ID: <88781f4075c14925bead0f7e68e20f11@samsung.com> [https://d24cgw3uvb9a9h.cloudfront.net/static/94026/image/new/ZoomLogo_110_25.png] Hi there, Bill Martin (Samsung) is inviting you to a scheduled Zoom meeting. Join Zoom Meeting Phone one-tap: US: +16699009128,,99403434298#,,,,,,0#,,49645580# or +12133388477,,99403434298#,,,,,,0#,,49645580# Meeting URL: https://samsung-dsa.zoom.us/j/99403434298?pwd=eTdYeC9SQWY0Sk1La2M4Z0hJSHpRZz09 Meeting ID: 994 0343 4298 Passcode: 4i.wejny Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 669 900 9128 or +1 213 338 8477 or +1 669 219 2599 or +1 602 753 0140 or +1 720 928 9299 or +1 971 247 1195 or +1 253 215 8782 or +1 346 248 7799 or +1 786 635 1003 or +1 267 831 0333 or +1 301 715 8592 or +1 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 or +1 646 518 9805 or +1 646 558 8656 or +1 651 372 8299 Meeting ID: 994 0343 4298 Passcode: 49645580 International numbers Join from an H.323/SIP room system H.323: 162.255.37.11 (US West) 162.255.36.11 (US East) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 994 0343 4298 Passcode: 49645580 SIP: 99403434298 at zoomcrc.com Passcode: 49645580 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3302 bytes Desc: not available URL: From bill.martin at samsung.com Fri Jul 24 09:50:07 2020 From: bill.martin at samsung.com (Bill Martin) Date: Fri, 24 Jul 2020 16:50:07 +0000 Subject: [T10] T10 plenary meeting #159 References: Message-ID: <85584f627c1c408394ab60e885d9be4d@samsung.com> [https://d24cgw3uvb9a9h.cloudfront.net/static/94026/image/new/ZoomLogo_110_25.png] Hi there, Bill Martin (Samsung) is inviting you to a scheduled Zoom meeting. Join Zoom Meeting Phone one-tap: US: +16699009128,,99592531267#,,,,,,0#,,04876961# or +12133388477,,99592531267#,,,,,,0#,,04876961# Meeting URL: https://samsung-dsa.zoom.us/j/99592531267?pwd=K3NLNG5ROHo2SzRwWVcraU53RzFUdz09 Meeting ID: 995 9253 1267 Passcode: 7j?H+9&3 Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 669 900 9128 or +1 213 338 8477 or +1 669 219 2599 or +1 720 928 9299 or +1 971 247 1195 or +1 253 215 8782 or +1 346 248 7799 or +1 602 753 0140 or +1 646 518 9805 or +1 646 558 8656 or +1 651 372 8299 or +1 786 635 1003 or +1 267 831 0333 or +1 301 715 8592 or +1 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 Meeting ID: 995 9253 1267 Passcode: 04876961 International numbers Join from an H.323/SIP room system H.323: 162.255.37.11 (US West) 162.255.36.11 (US East) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 995 9253 1267 Passcode: 04876961 SIP: 99592531267 at zoomcrc.com Passcode: 04876961 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3248 bytes Desc: not available URL: From DoNotReply at t10.org Sat Jul 25 00:02:23 2020 From: DoNotReply at t10.org (DoNotReply at t10.org) Date: Sat, 25 Jul 2020 00:02:23 -0700 Subject: [T10] Recent T10 documents uploaded since 2020/07/18 Message-ID: <01d66251$Blat.v2.2.2$8d17e29f@wdc.com> Proposals --------- UAS-3: Align transport protocol services with SAM-6 (by: Ralph Weber) T10/20-051r2 Uploaded: 2020/07/22 255532 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-051r2.pdf SPC-6 Taking time to ready back to the information field (by: Curtis Ballard) T10/20-061r1 Uploaded: 2020/07/18 955325 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-061r1.pdf SPC-6 Taking time to ready back to the information field (by: Curtis Ballard) T10/20-061r2 Uploaded: 2020/07/20 955833 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-061r2.pdf UAS-3: ERDY Clarifications (by: Ralph Weber) T10/20-066r1 Uploaded: 2020/07/24 73409 bytes https://www.t10.org/cgi-bin/ac.pl?t=d&f=20-066r1.pdf Working Drafts -------------- (Report generated on 2020/07/25 at 00:02:03) From Ralph.Weber at wdc.com Sat Jul 25 14:52:47 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Sat, 25 Jul 2020 21:52:47 +0000 Subject: [T10] T10.org network updates are finished Message-ID: The WD firewall network updates affecting t10.org have been finished. I've tried a few of my favorite incantations with excellent, expected results. All the best, .Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Wed Jul 29 06:47:56 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Wed, 29 Jul 2020 13:47:56 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: <53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> <53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> Message-ID: Proposal 20-082r0 SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From: t10-bounces at t10.org On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org >; t10-bounces at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kdbutt at us.ibm.com Wed Jul 29 08:48:45 2020 From: kdbutt at us.ibm.com (Kevin D Butt) Date: Wed, 29 Jul 2020 08:48:45 -0700 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> Message-ID: Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMB should be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODE field (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Proposal 20-082r0 SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From: t10-bounces at t10.org On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: on behalf of Mike Webster < Mike.Webster at wdc.com> Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: on behalf of Curtis Stevens < curtis.stevens at seagate.com> Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt , Paul Suhler Cc: "t10 at t10.org" , "t10-bounces at t10.org" < t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From: t10-bounces at t10.org on behalf of Kevin D Butt Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org ; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://urldefense.proofpoint.com/v2/url?u=https-3A__www.t10.org_mailman_listinfo_t10&d=DwICAg&c=jf_iaSHvJObTbx-siA1ZOg&r=9hRV7-xktvhH6SmVBcFXOfc3l09GaQd0rqZQQg8BGkY&m=GXNfhDunnLQbNt6cSdHaujI6MU24clCwXv-ldbVWJyQ&s=ZUDGecrmGl8y1ZdT5oGVTfDkuZ_mhEfID_Dk9w8hwzo&e= -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Wed Jul 29 12:03:51 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Wed, 29 Jul 2020 19:03:51 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> , Message-ID: I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Kevin D Butt Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.ballard at hpe.com Wed Jul 29 12:49:25 2020 From: curtis.ballard at hpe.com (Ballard, Curtis C (HPE Storage)) Date: Wed, 29 Jul 2020 19:49:25 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> , Message-ID: Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removable able to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Wed Jul 29 13:12:04 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Wed, 29 Jul 2020 20:12:04 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> , , Message-ID: I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: Ballard, Curtis C (HPE Storage) Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens ; Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removable able to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Wed Jul 29 13:19:34 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Wed, 29 Jul 2020 20:19:34 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> , , Message-ID: Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. From: Curtis Stevens Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) ; Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org > Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removable able to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Wed Jul 29 13:35:20 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Wed, 29 Jul 2020 20:35:20 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> , , , Message-ID: I like that one better. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Ralph Weber Sent: Wednesday, July 29, 2020 1:19 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. From: Curtis Stevens Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) ; Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org > Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removable able to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kdbutt at us.ibm.com Wed Jul 29 13:41:50 2020 From: kdbutt at us.ibm.com (Kevin D Butt) Date: Wed, 29 Jul 2020 20:41:50 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. From: Curtis Stevens Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) ; Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From: Ballard, Curtis C (HPE Storage) Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens ; Kevin D Butt < kdbutt at us.ibm.com>; Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removable able to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From: t10-bounces at t10.org on behalf of Kevin D Butt Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: on behalf of Mike Webster < Mike.Webster at wdc.com> Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: on behalf of Curtis Stevens < curtis.stevens at seagate.com> Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt , Paul Suhler Cc: "t10 at t10.org" , "t10-bounces at t10.org" < t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:t10-bounces at t10.org on behalf of Kevin D Butt < kdbutt at us.ibm.com> Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://urldefense.proofpoint.com/v2/url?u=https-3A__www.t10.org_mailman_listinfo_t10&d=DwICAg&c=jf_iaSHvJObTbx-siA1ZOg&r=9hRV7-xktvhH6SmVBcFXOfc3l09GaQd0rqZQQg8BGkY&m=Z0FcAH4JnCOKv_7ThAO1DVnc8TZIMU9PzGZalkvK1yA&s=UwQL32qBxqJ78U4lFYSdMOOWuQ8ExU3bNuHeyCebsYg&e= -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.ballard at hpe.com Wed Jul 29 13:43:18 2020 From: curtis.ballard at hpe.com (Ballard, Curtis C (HPE Storage)) Date: Wed, 29 Jul 2020 20:43:18 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> , , , Message-ID: As do I. Ralphs version makes better use of a base SCSI architecture concept that should be easily grasped. Removing the I_T nexus is the way the device server would be removed. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 2:35 PM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I like that one better. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Ralph Weber > Sent: Wednesday, July 29, 2020 1:19 PM To: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. From: Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org > Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removable able to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Wed Jul 29 17:51:37 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 00:51:37 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From John.Geldman at kioxia.com Wed Jul 29 18:26:56 2020 From: John.Geldman at kioxia.com (John Geldman) Date: Thu, 30 Jul 2020 01:26:56 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Wed Jul 29 18:45:27 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 01:45:27 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Mike.Webster at wdc.com Wed Jul 29 19:00:39 2020 From: Mike.Webster at wdc.com (Michael Webster) Date: Thu, 30 Jul 2020 02:00:39 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> <53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> Message-ID: Another suggestion: A removable medium (rmb) bit set to zero indicates that the medium is not removable must be present when the device server is connected to an application client through an I_T nexus. A rmb bit set to one indicates that the medium is removablemay be present when the device server is connected to an application client through an I_T nexus. Mike Webster Western Digital From: on behalf of Ralph Weber Date: Wednesday, July 29, 2020 at 6:46 PM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Mike.Webster at wdc.com Wed Jul 29 19:05:00 2020 From: Mike.Webster at wdc.com (Michael Webster) Date: Thu, 30 Jul 2020 02:05:00 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> <53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> Message-ID: <7EB7970A-24F5-455C-9B91-34F4F8033C74@wdc.com> Actually, I take that back. That ?may? statement would give license to USB sticks to keep doing what they are doing. Mike Webster Western Digital From: on behalf of Mike Webster Date: Wednesday, July 29, 2020 at 7:01 PM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Another suggestion: A removable medium (rmb) bit set to zero indicates that the medium is not removable must be present when the device server is connected to an application client through an I_T nexus. A rmb bit set to one indicates that the medium is removablemay be present when the device server is connected to an application client through an I_T nexus. Mike Webster Western Digital From: on behalf of Ralph Weber Date: Wednesday, July 29, 2020 at 6:46 PM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Mike.Webster at wdc.com Wed Jul 29 19:17:28 2020 From: Mike.Webster at wdc.com (Michael Webster) Date: Thu, 30 Jul 2020 02:17:28 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: <7EB7970A-24F5-455C-9B91-34F4F8033C74@wdc.com> References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com> <53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com> <7EB7970A-24F5-455C-9B91-34F4F8033C74@wdc.com> Message-ID: <34169C8E-3B47-48C4-9C10-2D0A70368736@wdc.com> Maybe the following is too strong, but aren?t we trying to prohibit the USB sticks from reporting RMB=1? A device server that can maintain an I_T nexus to an application client without its medium present may report RMB=1. A device server that cannot maintain an I_T nexus to an application client without its medium present must report RMB=0. Mike Webster Western Digital From: on behalf of Mike Webster Date: Wednesday, July 29, 2020 at 7:06 PM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Actually, I take that back. That ?may? statement would give license to USB sticks to keep doing what they are doing. Mike Webster Western Digital From: on behalf of Mike Webster Date: Wednesday, July 29, 2020 at 7:01 PM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Another suggestion: A removable medium (rmb) bit set to zero indicates that the medium is not removable must be present when the device server is connected to an application client through an I_T nexus. A rmb bit set to one indicates that the medium is removablemay be present when the device server is connected to an application client through an I_T nexus. Mike Webster Western Digital From: on behalf of Ralph Weber Date: Wednesday, July 29, 2020 at 6:46 PM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From John.Geldman at kioxia.com Wed Jul 29 22:37:03 2020 From: John.Geldman at kioxia.com (John Geldman) Date: Thu, 30 Jul 2020 05:37:03 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: <87cf48d479784ffeb2dab63dca17d93c@kioxia.com> Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Thu Jul 30 04:34:36 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 11:34:36 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: <87cf48d479784ffeb2dab63dca17d93c@kioxia.com> References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , <87cf48d479784ffeb2dab63dca17d93c@kioxia.com> Message-ID: Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Thu Jul 30 07:56:08 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Thu, 30 Jul 2020 14:56:08 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , <87cf48d479784ffeb2dab63dca17d93c@kioxia.com>, Message-ID: Curiosity question... Has anyone found why this practice is happening and/or being requested?? Just wondering if Windows or iOS provide a different experience when the bit is set? --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Ralph Weber Sent: Thursday, July 30, 2020 4:34 AM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, Having slept on the problem, it?s become apparent that symmetry requires the addition of ?is not able to be? to the first sentence, a la? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn?t the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.stevens at seagate.com Thu Jul 30 08:15:25 2020 From: curtis.stevens at seagate.com (Curtis Stevens) Date: Thu, 30 Jul 2020 15:15:25 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , <87cf48d479784ffeb2dab63dca17d93c@kioxia.com>, , Message-ID: The reason I ask this question is that in windows, when you right click the drive letter in windows explorer, some devices show eject in that menu, others do not. It is fewer keystrokes and targets the correct device more effectivly than going into the task bar, left clicking the uparrow, left clicking mouseman and then selecting the device. I actually think the standard is fine, if a little vague, but it really the OS behavior and user convenience that is driving the issue. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Curtis Stevens Sent: Thursday, July 30, 2020 7:56 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Curiosity question... Has anyone found why this practice is happening and/or being requested?? Just wondering if Windows or iOS provide a different experience when the bit is set? --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org on behalf of Ralph Weber Sent: Thursday, July 30, 2020 4:34 AM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, Having slept on the problem, it?s become apparent that symmetry requires the addition of ?is not able to be? to the first sentence, a la? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn?t the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kdbutt at us.ibm.com Thu Jul 30 08:41:43 2020 From: kdbutt at us.ibm.com (Kevin D Butt) Date: Thu, 30 Jul 2020 15:41:43 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Kevin, Having slept on the problem, it?s become apparent that symmetry requires the addition of ?is not able to be? to the first sentence, a la? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn?t the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) ; Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:Ballard, Curtis C (HPE Storage) Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens ; Kevin D Butt < kdbutt at us.ibm.com>; Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:t10-bounces at t10.org on behalf of Kevin D Butt < kdbutt at us.ibm.com> Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: on behalf of Mike Webster < Mike.Webster at wdc.com> Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: on behalf of Curtis Stevens < curtis.stevens at seagate.com> Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt , Paul Suhler Cc: "t10 at t10.org" , "t10-bounces at t10.org" < t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:t10-bounces at t10.org on behalf of Kevin D Butt < kdbutt at us.ibm.com> Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://urldefense.proofpoint.com/v2/url?u=https-3A__www.t10.org_mailman_listinfo_t10&d=DwICAg&c=jf_iaSHvJObTbx-siA1ZOg&r=9hRV7-xktvhH6SmVBcFXOfc3l09GaQd0rqZQQg8BGkY&m=O0q3G8h13e08pkqZ5JYhkfyFUnqkgi9A101sS7dM5FM&s=qfrJV1YJbsMF_WY8Rzj0HZpNkDlfOeNkWpDLAWgovMs&e= -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Thu Jul 30 08:43:46 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 15:43:46 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , <87cf48d479784ffeb2dab63dca17d93c@kioxia.com>, Message-ID: Yes, I have WD Confidential knowledge of the motivation for this proposal. From: Curtis Stevens Sent: Thursday, July 30, 2020 9:56 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Curiosity question... Has anyone found why this practice is happening and/or being requested?? Just wondering if Windows or iOS provide a different experience when the bit is set? --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Ralph Weber > Sent: Thursday, July 30, 2020 4:34 AM To: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Thu Jul 30 08:54:46 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 15:54:46 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , <87cf48d479784ffeb2dab63dca17d93c@kioxia.com>, , Message-ID: IMHO The discussion below appears to be covering the kind of material that would cause a T10 Plenary to be recessed if it were mentioned in that forum. From: Curtis Stevens Sent: Thursday, July 30, 2020 10:15 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The reason I ask this question is that in windows, when you right click the drive letter in windows explorer, some devices show eject in that menu, others do not. It is fewer keystrokes and targets the correct device more effectivly than going into the task bar, left clicking the uparrow, left clicking mouseman and then selecting the device. I actually think the standard is fine, if a little vague, but it really the OS behavior and user convenience that is driving the issue. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Curtis Stevens > Sent: Thursday, July 30, 2020 7:56 AM To: Ralph Weber >; t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Curiosity question... Has anyone found why this practice is happening and/or being requested?? Just wondering if Windows or iOS provide a different experience when the bit is set? --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Ralph Weber > Sent: Thursday, July 30, 2020 4:34 AM To: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Thu Jul 30 11:58:41 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 18:58:41 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: Kevin, This doesn't look too difficult to solve. A removable medium (rmb) bit set to zero indicates that the medium is not removable or is able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: Kevin D Butt Sent: Thursday, July 30, 2020 10:42 AM To: Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From:John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From:Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kdbutt at us.ibm.com Thu Jul 30 12:08:28 2020 From: kdbutt at us.ibm.com (Kevin D Butt) Date: Thu, 30 Jul 2020 19:08:28 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: Ralph, This looks good to me. I am still trying to wrap my head around "break the I_T nexus" since this is the first use in SPC or SAM. However, I think the standard English meaning works well here. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/30/2020 11:59 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Kevin, This doesn?t look too difficult to solve. A removable medium (rmb) bit set to zero indicates that the medium is not removable or is able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: Kevin D Butt Sent: Thursday, July 30, 2020 10:42 AM To: Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Kevin, Having slept on the problem, it?s become apparent that symmetry requires the addition of ?is not able to be? to the first sentence, a la? A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:John Geldman Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn?t the problem folks thinking RMB was/is about removable devices? John From:t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From:John Geldman Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From:t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however? I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From:Kevin D Butt Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Or? how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) ; Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:Ballard, Curtis C (HPE Storage) Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens ; Kevin D Butt < kdbutt at us.ibm.com>; Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don?t we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt ; Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:t10-bounces at t10.org on behalf of Kevin D Butt < kdbutt at us.ibm.com> Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber To: "t10 at t10.org" Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a ?real? removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: on behalf of Mike Webster < Mike.Webster at wdc.com> Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system?s driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system?s driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: on behalf of Curtis Stevens < curtis.stevens at seagate.com> Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt , Paul Suhler Cc: "t10 at t10.org" , "t10-bounces at t10.org" < t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 From:t10-bounces at t10.org on behalf of Kevin D Butt < kdbutt at us.ibm.com> Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler Cc: t10 at t10.org; t10-bounces at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler To: "Ballard, Curtis C (HPE Storage)" , Ralph Weber , "t10 at t10.org" Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber ; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I?ve used with that bit set to one, and I?ve worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable.? An interesting debate has developed regarding whether a USB ?memory stick? is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of? ?A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition.? How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://urldefense.proofpoint.com/v2/url?u=https-3A__www.t10.org_mailman_listinfo_t10&d=DwICAg&c=jf_iaSHvJObTbx-siA1ZOg&r=9hRV7-xktvhH6SmVBcFXOfc3l09GaQd0rqZQQg8BGkY&m=Om72fFJOH6-mMf4MdNjQwQrDxd6cPhaqNd9iFJam1TM&s=n87abEwkItoStZjQwHExKHJ1pMg_6D3-lDP_5dM_rzI&e= -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Thu Jul 30 13:12:33 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Thu, 30 Jul 2020 20:12:33 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: Kevin, The problem is that SAM has little, if any, idea of I_T nexuses coming or going. I_T nexuses simply are there (most of the time). There is the I_T NEXUS RESET task management function, but it's not about vaporizing an I_T nexus. In the Events Model, there is a Power On, but that's about things starting. There is Power Loss Expected, but that is not associated with a guarantee of power being lost. I'd like to avoid using the verb "to remove" relative to I_T nexuses, so that "to remove" can be associated solely with the medium. Probably, I should leave "breaking" and let CAP Zen it's way to a widely agreeable replacement. All the best, .Ralph From: Kevin D Butt Sent: Thursday, July 30, 2020 2:08 PM To: Ralph Weber Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, This looks good to me. I am still trying to wrap my head around "break the I_T nexus" since this is the first use in SPC or SAM. However, I think the standard English meaning works well here. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 11:59 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, This doesn't look too difficult to solve. A removable medium (rmb) bit set to zero indicates that the medium is not removable or is able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:Kevin D Butt > Sent: Thursday, July 30, 2020 10:42 AM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From:John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From:Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From John.Geldman at kioxia.com Thu Jul 30 13:11:20 2020 From: John.Geldman at kioxia.com (John Geldman) Date: Thu, 30 Jul 2020 20:11:20 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , <87cf48d479784ffeb2dab63dca17d93c@kioxia.com>, , Message-ID: I will note that when it comes to user interfaces, what people assume tends to be a higher value suit of cards over our obtuse and obfuscated specifications (i.e., hearts beat clubs). Confirming what multiple OS's historically did with the bit and what we initially meant by the bit should inform our next steps for clarifying interoperability intent. John From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 30, 2020 8:55 AM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The discussion below appears to be covering the kind of material that would cause a T10 Plenary to be recessed if it were mentioned in that forum. From: Curtis Stevens > Sent: Thursday, July 30, 2020 10:15 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The reason I ask this question is that in windows, when you right click the drive letter in windows explorer, some devices show eject in that menu, others do not. It is fewer keystrokes and targets the correct device more effectivly than going into the task bar, left clicking the uparrow, left clicking mouseman and then selecting the device. I actually think the standard is fine, if a little vague, but it really the OS behavior and user convenience that is driving the issue. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Curtis Stevens > Sent: Thursday, July 30, 2020 7:56 AM To: Ralph Weber >; t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Curiosity question... Has anyone found why this practice is happening and/or being requested?? Just wondering if Windows or iOS provide a different experience when the bit is set? --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Ralph Weber > Sent: Thursday, July 30, 2020 4:34 AM To: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From curtis.ballard at hpe.com Thu Jul 30 13:49:03 2020 From: curtis.ballard at hpe.com (Ballard, Curtis C (HPE Storage)) Date: Thu, 30 Jul 2020 20:49:03 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: Why isn't an I_T nexus removal equivalent to an I_T nexus loss? I can't think of any SAM term about establishing an I_T nexus but it does talk about I_T nexuses going away as an I_T nexus loss. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Ralph Weber Sent: Thursday, July 30, 2020 2:13 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, The problem is that SAM has little, if any, idea of I_T nexuses coming or going. I_T nexuses simply are there (most of the time). There is the I_T NEXUS RESET task management function, but it's not about vaporizing an I_T nexus. In the Events Model, there is a Power On, but that's about things starting. There is Power Loss Expected, but that is not associated with a guarantee of power being lost. I'd like to avoid using the verb "to remove" relative to I_T nexuses, so that "to remove" can be associated solely with the medium. Probably, I should leave "breaking" and let CAP Zen it's way to a widely agreeable replacement. All the best, .Ralph From: Kevin D Butt > Sent: Thursday, July 30, 2020 2:08 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, This looks good to me. I am still trying to wrap my head around "break the I_T nexus" since this is the first use in SPC or SAM. However, I think the standard English meaning works well here. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 11:59 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, This doesn't look too difficult to solve. A removable medium (rmb) bit set to zero indicates that the medium is not removable or is able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:Kevin D Butt > Sent: Thursday, July 30, 2020 10:42 AM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From:John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From:Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From John.Geldman at kioxia.com Thu Jul 30 17:14:07 2020 From: John.Geldman at kioxia.com (John Geldman) Date: Fri, 31 Jul 2020 00:14:07 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: <4f418b5c39a04ae18febe8d7d935f229@kioxia.com> Some perspective please: is the thought to alter SPC for greater device developer understanding, or SAM for advanced students of the SCSI way? From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 30, 2020 1:13 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, The problem is that SAM has little, if any, idea of I_T nexuses coming or going. I_T nexuses simply are there (most of the time). There is the I_T NEXUS RESET task management function, but it's not about vaporizing an I_T nexus. In the Events Model, there is a Power On, but that's about things starting. There is Power Loss Expected, but that is not associated with a guarantee of power being lost. I'd like to avoid using the verb "to remove" relative to I_T nexuses, so that "to remove" can be associated solely with the medium. Probably, I should leave "breaking" and let CAP Zen it's way to a widely agreeable replacement. All the best, .Ralph From: Kevin D Butt > Sent: Thursday, July 30, 2020 2:08 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, This looks good to me. I am still trying to wrap my head around "break the I_T nexus" since this is the first use in SPC or SAM. However, I think the standard English meaning works well here. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 11:59 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, This doesn't look too difficult to solve. A removable medium (rmb) bit set to zero indicates that the medium is not removable or is able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:Kevin D Butt > Sent: Thursday, July 30, 2020 10:42 AM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From:John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From:Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Fri Jul 31 05:44:46 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Fri, 31 Jul 2020 12:44:46 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <37154D0A-396A-4CA7-9148-BCBBF94D83D9@wdc.com><53C9BD00-40D2-4961-B7B5-8803A50F1F41@wdc.com>, , Message-ID: This is very embarrassing. I looked SAM-6 events figure and somehow totally ignored the I_T nexus loss event. I'll have to claim a "senior moment" for that one. Since "I_T nexus loss" is mentioned over 50 times in SAM-6, there can be no denying that the topic is solidly covered in the Architecture Model. With this brick neatly installed in the foundation, I have posted 20-082r1 to reflect the sum of the detailed corrections discussed on this thread. Enjoy! Thanks to Curtis Ballard and Kevin Butt for their unflagging efforts to make 20-082 a best-in-class proposal. All the best, .Ralph From: Ballard, Curtis C (HPE Storage) Sent: Thursday, July 30, 2020 3:49 PM To: Ralph Weber ; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Why isn't an I_T nexus removal equivalent to an I_T nexus loss? I can't think of any SAM term about establishing an I_T nexus but it does talk about I_T nexuses going away as an I_T nexus loss. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Ralph Weber Sent: Thursday, July 30, 2020 2:13 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, The problem is that SAM has little, if any, idea of I_T nexuses coming or going. I_T nexuses simply are there (most of the time). There is the I_T NEXUS RESET task management function, but it's not about vaporizing an I_T nexus. In the Events Model, there is a Power On, but that's about things starting. There is Power Loss Expected, but that is not associated with a guarantee of power being lost. I'd like to avoid using the verb "to remove" relative to I_T nexuses, so that "to remove" can be associated solely with the medium. Probably, I should leave "breaking" and let CAP Zen it's way to a widely agreeable replacement. All the best, .Ralph From: Kevin D Butt > Sent: Thursday, July 30, 2020 2:08 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, This looks good to me. I am still trying to wrap my head around "break the I_T nexus" since this is the first use in SPC or SAM. However, I think the standard English meaning works well here. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 11:59 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, This doesn't look too difficult to solve. A removable medium (rmb) bit set to zero indicates that the medium is not removable or is able to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:Kevin D Butt > Sent: Thursday, July 30, 2020 10:42 AM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, That does solve the mentioned problems. It does, however, cause a nuanced change in meaning from "not removable" to "able to be removed only if". The original definition says if the RMB bit is zero the medium is not removable whereas the new definition says the medium can actually be removed, but only if the device server goes away with it. While that does seem to be the intent of this effort, I have to wonder if it is not actually changing the meaning. My concern is rooted in making sure we don't break something else. From tape devices, I do not see the proposed definition to be an issue. However, I don't have much a view into the application client side of things outside the tape world. Do application clients expect that the medium is never removed is the RMB bit is set to zero? If the I_T nexus is broken, do they search other I_T nexuses to try and find the medium? Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/30/2020 04:35 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From:John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From:John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From:Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From tjmac at tolisgroup.com Fri Jul 31 08:50:01 2020 From: tjmac at tolisgroup.com (Tim Jones) Date: Fri, 31 Jul 2020 08:50:01 -0700 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: References: <87cf48d479784ffeb2dab63dca17d93c@kioxia.com> Message-ID: <5ed78b9c-5e03-1adf-edd0-9d1b644b86a0@tolisgroup.com> Hi all, If I may interject as an ISV and direct interface to said "hearts", when software deals with removal, it's not considering that you removed a - for example - physical eSATA disk chassis, but that the filesystem provided by that chassis was removed, or the tape cartridge was ejected from the tape drive.? Users, therefore, are less aware of the device versus media conundrum that this discussion considers.? Since the RMB entity is something that developers need, but users never see directly, I would "vote" for not clumping the ability to identify a device as removable with identifying media removability.? Physical devices should be identified by their "hot plug" ability, which is very different. Tim Jones TOLIS Group, Inc. On 7/30/20 1:11 PM, John Geldman wrote: > > I will note that when it comes to user interfaces, what people assume > tends to be a higher value suit of cards over our obtuse and > obfuscated specifications (i.e., hearts beat clubs). > > Confirming what multiple OS?s historically did with the bit and what > we initially meant by the bit should inform our next steps for > clarifying interoperability intent. > > John > > *From:* t10-bounces at t10.org *On Behalf Of *Ralph > Weber > *Sent:* Thursday, July 30, 2020 8:55 AM > *To:* t10 at t10.org > *Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > IMHO The discussion below appears to be covering the kind of material > that would cause a T10 Plenary to be recessed if it were mentioned in > that forum. > > *From:* Curtis Stevens > > *Sent:* Thursday, July 30, 2020 10:15 AM > *To:* Ralph Weber >; > t10 at t10.org > *Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > *CAUTION:**This email originated from outside of Western Digital. Do > not click on links or open attachments unless you recognize the sender > and know that the content is safe.* > > The reason I ask this question is that in windows, when you right > click the drive letter in windows explorer, some devices show eject in > that menu, others do not.? It is fewer keystrokes and targets the > correct device more effectivly than going into the task bar, left > clicking the uparrow, left clicking mouseman and then selecting the > device. > > I actually think the standard is fine, if a little vague, but it > really the OS behavior and user convenience that is driving the issue. > > --------------------------------------------- > > Curtis E. Stevens > > Technologist > > Seagate Technology > > E-Mail: Curtis.Stevens at Seagate.com > > Phone: 949-307-5050 > > ------------------------------------------------------------------------ > > *From:*t10-bounces at t10.org > > on behalf of Curtis > Stevens > > *Sent:* Thursday, July 30, 2020 7:56 AM > *To:* Ralph Weber >; > t10 at t10.org > > *Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > Curiosity question...? Has anyone found why this practice is happening > and/or being requested?? > > Just wondering if Windows or iOS provide a different experience when > the bit is set? > > --------------------------------------------- > > Curtis E. Stevens > > Technologist > > Seagate Technology > > E-Mail: Curtis.Stevens at Seagate.com > > Phone: 949-307-5050 > > ------------------------------------------------------------------------ > > *From:*t10-bounces at t10.org > > on behalf of Ralph > Weber > > *Sent:* Thursday, July 30, 2020 4:34 AM > *To:* t10 at t10.org > > *Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > Kevin, > > Having slept on the problem, it?s become apparent that symmetry > requires the addition of ?is not able to be? to the first sentence, a la? > > A removable medium (rmb) bit set to zero indicates that the medium is > not removableable to be removed only if that removal is accompanied by > the breaking of the I_T nexus connection between the application > client to the device server. A rmb bit set to one indicates that the > medium is removableable to be removed without affecting the I_T nexus > between the application client and the device server. > > All the best, .Ralph > > *From:* John Geldman > > *Sent:* Thursday, July 30, 2020 12:37 AM > *To:* Ralph Weber >; > t10 at t10.org > *Subject:* RE: [T10] Revisiting the RMB (Removable Medium) bit definition > > *CAUTION:**This email originated from outside of Western Digital. Do > not click on links or open attachments unless you recognize the sender > and know that the content is safe.* > > Ralph, > > I seem to be missing something. > > From my world view RMB was/is about removable media, not removable > devices. We still want it to be. If we declare that rmb set to zero > means not able to remove the media, then to remove the media, the > device server has to be removed, i.e., disconnected. > > The statement that removal of a device breaks the connection to the > host needs to be said? > > Isn?t the problem folks thinking RMB was/is about removable devices? > > John > > *From:* t10-bounces at t10.org > > *On Behalf Of > *Ralph Weber > *Sent:* Wednesday, July 29, 2020 6:45 PM > *To:* t10 at t10.org > *Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > IMHO The absence of any statement about the I_T nexus in the RMB=0 > case severely limits the value of the change. > > *From:* John Geldman > > *Sent:* Wednesday, July 29, 2020 8:27 PM > *To:* Ralph Weber >; > t10 at t10.org > *Subject:* RE: [T10] Revisiting the RMB (Removable Medium) bit definition > > *CAUTION:**This email originated from outside of Western Digital. Do > not click on links or open attachments unless you recognize the sender > and know that the content is safe.* > > The observations are all correct, but they do obfuscate. Does this work? > > A removable medium (rmb) bit set to zero indicates that the medium is > not removable able to be removed from the device server. A rmb bit set > to one indicates that the medium is removableable to be removed from > the device server. If the rmb bit is set to one and the medium is > removed, then the I_T nexus between the application client and the > device server, if any, is not affected. > > John > > *From:* t10-bounces at t10.org > > *On Behalf Of > *Ralph Weber > *Sent:* Wednesday, July 29, 2020 5:52 PM > *To:* t10 at t10.org > *Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > I agree that the two sentence are not symmetrical, however? I cannot > find a way to make them symmetrical that does not involve an excess of > Rube Goldberg English that obfuscates more than it clarifies. > > Please feel free to augment the suggestion made below with detailed > text changes. > > All the best, .Ralph > > *From:* Kevin D Butt > > *Sent:* Wednesday, July 29, 2020 3:42 PM > *To:* Ralph Weber > > *Cc:* t10 at t10.org > *Subject:* RE: [T10] Revisiting the RMB (Removable Medium) bit definition > > *CAUTION:**This email originated from outside of Western Digital. Do > not click on links or open attachments unless you recognize the sender > and know that the content is safe.* > > That is starting to look good to me, but I would suggest making it > symmetrical. > > A removable medium (rmb) bit set to zero indicates that the medium is > not removable able to be removed without removing the I_T nexus that > connects the application client to the device server. A rmb bit set to > one indicates that the medium is removableable to be removed without > affecting the I_T nexus between the application client and the device > server. > > > > Kevin D. Butt > SCSI Architect, Tape Firmware, Data Retention Infrastructure > T10 Standards > MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 > Tel: 520-799-5280 > Fax: 520-799-2723 (T/L:321) > Email address: kdbutt at us.ibm.com > > =========== Interesting Links =========== > [ IBM Tape Storage ] > https://www.ibm.com/it-infrastructure/storage/tape > > [ SSIC - HBA/OS/Switch/Product interoperation ] > https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss > > [ LTO & 3592 ISV Support Matrix ] > www.ibm.com/systems/resources/lto_isv_matrix.pdf > > [ LTO SCSI Reference ] > http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 > > [ 3592 SCSI Reference ] > http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 > > =================================== > > > > From: Ralph Weber > > To: "t10 at t10.org " > > Date: 07/29/2020 13:20 > Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) > bit definition > Sent by: t10-bounces at t10.org > > ------------------------------------------------------------------------ > > Or? how about something along the lines of: > > A removable medium (rmb) bit set to zero indicates that the medium is > not removable without removing the I_T nexus that connects the > application client to the device server. A rmb bit set to one > indicates that the medium is removableable to be removed without > affecting the I_T nexus between the application client and the device > server. > > *From:*Curtis Stevens > * > Sent:* Wednesday, July 29, 2020 3:12 PM* > To:* Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber >* > Cc:* t10 at t10.org * > Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > *CAUTION:**This email originated from outside of Western Digital. Do > not click on links or open attachments unless you recognize the sender > and know that the content is safe.* > > I think part of the blatant wording that Ralph is proposing is that > those making the requirements on the device side do not fully > understand SCSI. ?This has been the case from day 1 and has not > changed. ?It is fairly easy to understand commands and what they do. > ?Understanding the architecture is a different story. > > I think your suggested change could be integrated with Ralphs change... > > --------------------------------------------- > > Curtis E. Stevens > > Technologist > > Seagate Technology > > E-Mail: Curtis.Stevens at Seagate.com > > Phone: 949-307-5050 > > ------------------------------------------------------------------------ > > *From:*Ballard, Curtis C (HPE Storage) >* > Sent:* Wednesday, July 29, 2020 12:49 PM* > To:* Curtis Stevens >; Kevin D Butt >; Ralph Weber >* > Cc:* t10 at t10.org >* > Subject:* RE: [T10] Revisiting the RMB (Removable Medium) bit definition > > Why don?t we specify it in architectural terms. ?Something along the > lines of: > > A removable medium (rmb) bit set to zero indicates that the medium is > not removable from the device server. > > A rmb bit set to one indicates that the medium is removableable to be > removed without removing the device server. > > Curtis Ballard > > Hewlett Packard Enterprise > > HPE Storage R&D > > Fort Collins, CO > > (970) 898-3013 > > *From:*t10-bounces at t10.org > [mailto:t10-bounces at t10.org] *On Behalf Of > *Curtis Stevens* > Sent:* Wednesday, July 29, 2020 1:04 PM* > To:* Kevin D Butt >; > Ralph Weber >* > Cc:* t10 at t10.org * > Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > I would turn the proposal around and say the RMB bit shall only be set > to one if the device is able to report ... after is media is ejected. > > --------------------------------------------- > > Curtis E. Stevens > > Technologist > > Seagate Technology > > E-Mail: Curtis.Stevens at Seagate.com > > Phone: 949-307-5050 > > ------------------------------------------------------------------------ > > *From:*t10-bounces at t10.org > > on behalf of Kevin D Butt > >* > Sent:* Wednesday, July 29, 2020 8:48 AM* > To:* Ralph Weber >* > Cc:* t10 at t10.org >* > Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > Ralph, > > I applaud your goal, however, I am concerned about this definition. > Take, for example, a tape library. It has removable media, yet it > would not report one of these two additional sense codes. Even if all > user cartridges were removed, the library would still be ready. It > could do management type things including using diagnostic cartridges > (not seen as medium to the application client) to test drives. > > The RMBshould be set to zero, if the device server is not able to > terminate a TEST UNIT READY command (6.48) with CHECK CONDITION > status, with the sense key set to NOT READY and the additional sense code: > a)set to NOT READY, MEDIUM NOT PRESENT; or > b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah > (e.g., MEDIUM NOT PRESENT - TRAY OPEN). > > Thanks, > > Kevin D. Butt > SCSI Architect, Tape Firmware, Data Retention Infrastructure > T10 Standards > MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 > Tel: 520-799-5280 > Fax: 520-799-2723 (T/L:321) > Email address: kdbutt at us.ibm.com > > =========== Interesting Links =========== > [ IBM Tape Storage ] > https://www.ibm.com/it-infrastructure/storage/tape > > [ SSIC - HBA/OS/Switch/Product interoperation ] > https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss > > [ LTO & 3592 ISV Support Matrix ] > www.ibm.com/systems/resources/lto_isv_matrix.pdf > > [ LTO SCSI Reference ] > http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 > > [ 3592 SCSI Reference ] > http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 > > =================================== > > > > From: Ralph Weber > > To: "t10 at t10.org " > > Date: 07/29/2020 06:49 > Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) > bit definition > Sent by: t10-bounces at t10.org > > ------------------------------------------------------------------------ > > > Proposal 20-082r0 > /SPC-6: > Removable Medium Bit Expectations/ has been posted to bring this > discussion thread to the attention of CAP. > > Resisting the ongoing calls to put as many words as possible into > every standard, the proposal includes only one of the methods for > defining a ?real? removable medium device discussed in this thread. > For the purposes discussed here, only one method is necessary to > sufficiently define the standard. > > All the best, .Ralph > > *From:*t10-bounces at t10.org > *On Behalf Of *Michael Webster* > Sent:* Tuesday, July 21, 2020 7:57 PM* > To:* t10 at t10.org * > Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > I would like to add one other tidbit: ?A device reporting RMB=1 in > addition to being capable of a state that reports MEDIUM NOT PRESENT, > such a device should also properly support and execute an ejection > operation of that media when sent a START/STOP UNIT command with > Power_Condition=0, LoEj=1, & Start=0. > > Mike Webster > Western Digital > > *From: *> on behalf > of Mike Webster >* > Date: *Friday, July 17, 2020 at 10:30 AM* > To: *"t10 at t10.org " >* > Subject: *Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > I would expect a USB memory stick to report RMB=0 and ejection causes > the whole device to be removed from the operating system?s driver > stack. ?Afterward, such an ejected device would require an > unplug/re-plug for the device to respond to anything on USB. > > Unless that USB memory stick included a slot for media (e.g. an SD > slot) then I would expect the USB memory stick to report RMB=1 and > ejection would only cause the device to remove the media, the device > would remain in the operating system?s driver stack, and the device > would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT > for TEST UNIT READY. > > Mike Webster > Western Digital > > *From: *> on behalf > of Curtis Stevens >* > Date: *Friday, July 17, 2020 at 10:05 AM* > To: *Kevin D Butt >, Paul > Suhler >* > Cc: *"t10 at t10.org " >, "t10-bounces at t10.org > " >* > Subject: *Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > We had this descussion during the original specification development. > ?For USB, the device controller is removed along with the media. > ?There is nothing present to respond with an RMB bit. ?If however, you > plug in a USB floppy, you can have an RMB bit that enables eject > function reporting. > > > --------------------------------------------- > Curtis E. Stevens > Technologist > Seagate Technology > > E-Mail: Curtis.Stevens at Seagate.com > Phone: 949-307-5050 > > ------------------------------------------------------------------------ > > > *From:*t10-bounces at t10.org > > on behalf of Kevin D Butt > >* > Sent:* Friday, July 17, 2020 9:40 AM* > To:* Paul Suhler >* > Cc:* t10 at t10.org >; t10-bounces at t10.org > >* > Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > I concur with Curtis and Paul. > > Kevin D. Butt > SCSI Architect, Tape Firmware, Data Retention Infrastructure > T10 Standards > MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 > Tel: 520-799-5280 > Fax: 520-799-2723 (T/L:321) > Email address: kdbutt at us.ibm.com > > =========== Interesting Links =========== > [ IBM Tape Storage ] > https://www.ibm.com/it-infrastructure/storage/tape > > [ SSIC - HBA/OS/Switch/Product interoperation ] > https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss > > [ LTO & 3592 ISV Support Matrix ] > www.ibm.com/systems/resources/lto_isv_matrix.pdf > > [ LTO SCSI Reference ] > http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 > > [ 3592 SCSI Reference ] > http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 > > =================================== > > > > From: Paul Suhler > > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org " > > > Date: 07/17/2020 09:19 > Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) > bit definition > Sent by: t10-bounces at t10.org > > ------------------------------------------------------------------------ > > > I agree with Curtis about the need to be able to report NOT READY, > MEDIUM NOT PRESENT. I?ve worked on both SSC and SBC devices like that. > > Paul > Kioxia > * > From:*t10-bounces at t10.org > *On Behalf Of *Ballard, Curtis C (HPE > Storage)* > Sent:* Friday, July 17, 2020 9:54 AM* > To:* Ralph Weber >; > t10 at t10.org * > Subject:* Re: [T10] Revisiting the RMB (Removable Medium) bit definition > > Ralph, > > That bit is used with tape drives and was used with SCSI magneto > optical drives, DVD drives, etc. > > Even system I?ve used with that bit set to one, and I?ve worked with > several devices from different manufacturers that set that bit to one, > only use it in reference to the media with no device server disruption > other than the expected Unit Attention condition transitions as the > media is removed and re-loaded. > > SCSI devices should only set that bit to one if they are able to > report an 02h/3Ah/00h ? NOT READY, MEDIUM NOT PRESENT > > Curtis Ballard > Hewlett Packard Enterprise > * > From:*t10-bounces at t10.org > > *On Behalf Of *Ralph Weber* > Sent:* Thursday, July 16, 2020 5:27 PM* > To:* t10 at t10.org * > Subject:* [T10] Revisiting the RMB (Removable Medium) bit definition > > Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says? > ?A removable medium (RMB) bit set to zero indicates that the medium is > not removable. A RMB bit set to one indicates that the medium is > removable.? > > An interesting debate has developed regarding whether a USB ?memory > stick? is an RMB=0 or an RMB=1 device. > > One wag has suggested that the RMB bit definition be conditionalized > on the presence/absence of a coincident Logical Unit Reset condition > or I_T Nexus Loss condition, something along the lines of? > ?A removable medium (RMB) bit set to zero indicates that the medium is > not removable _without resulting in a Logical Unit Reset condition > (see SAM-6) or an I_T Nexus Loss condition (see SAM-6)_. A RMB bit set > to one indicates that the medium is removable _with no concurrent > Logical Unit Reset condition or I_T Nexus Loss condition_.? > > How says the T10 body politic? > > Thanks, > > .Ralph_______________________________________________ > T10 mailing list_ > _T10 at t10.org _ > _https://www.t10.org/mailman/listinfo/t10 > > > > _______________________________________________ > T10 mailing list_ > _T10 at t10.org _ > _https://www.t10.org/mailman/listinfo/t10 > > > > _______________________________________________ > T10 mailing list > T10 at t10.org > https://www.t10.org/mailman/listinfo/t10 > > > > _______________________________________________ > T10 mailing list > T10 at t10.org > https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ralph.Weber at wdc.com Fri Jul 31 09:46:52 2020 From: Ralph.Weber at wdc.com (Ralph Weber) Date: Fri, 31 Jul 2020 16:46:52 +0000 Subject: [T10] Revisiting the RMB (Removable Medium) bit definition In-Reply-To: <5ed78b9c-5e03-1adf-edd0-9d1b644b86a0@tolisgroup.com> References: <87cf48d479784ffeb2dab63dca17d93c@kioxia.com> <5ed78b9c-5e03-1adf-edd0-9d1b644b86a0@tolisgroup.com> Message-ID: As best as I can understand the "hearts" comment, it supports the changes proposed in 20-082r1. If this is not the case, I must rely on receiving a more specific analysis of the proposed text changes in the RMB bit definition to guide the next steps. All the best, .Ralph From: t10-bounces at t10.org On Behalf Of Tim Jones Sent: Friday, July 31, 2020 10:50 AM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Hi all, If I may interject as an ISV and direct interface to said "hearts", when software deals with removal, it's not considering that you removed a - for example - physical eSATA disk chassis, but that the filesystem provided by that chassis was removed, or the tape cartridge was ejected from the tape drive. Users, therefore, are less aware of the device versus media conundrum that this discussion considers. Since the RMB entity is something that developers need, but users never see directly, I would "vote" for not clumping the ability to identify a device as removable with identifying media removability. Physical devices should be identified by their "hot plug" ability, which is very different. Tim Jones TOLIS Group, Inc. On 7/30/20 1:11 PM, John Geldman wrote: I will note that when it comes to user interfaces, what people assume tends to be a higher value suit of cards over our obtuse and obfuscated specifications (i.e., hearts beat clubs). Confirming what multiple OS's historically did with the bit and what we initially meant by the bit should inform our next steps for clarifying interoperability intent. John From: t10-bounces at t10.org On Behalf Of Ralph Weber Sent: Thursday, July 30, 2020 8:55 AM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The discussion below appears to be covering the kind of material that would cause a T10 Plenary to be recessed if it were mentioned in that forum. From: Curtis Stevens > Sent: Thursday, July 30, 2020 10:15 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The reason I ask this question is that in windows, when you right click the drive letter in windows explorer, some devices show eject in that menu, others do not. It is fewer keystrokes and targets the correct device more effectivly than going into the task bar, left clicking the uparrow, left clicking mouseman and then selecting the device. I actually think the standard is fine, if a little vague, but it really the OS behavior and user convenience that is driving the issue. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Curtis Stevens > Sent: Thursday, July 30, 2020 7:56 AM To: Ralph Weber >; t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Curiosity question... Has anyone found why this practice is happening and/or being requested?? Just wondering if Windows or iOS provide a different experience when the bit is set? --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From: t10-bounces at t10.org > on behalf of Ralph Weber > Sent: Thursday, July 30, 2020 4:34 AM To: t10 at t10.org > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Kevin, Having slept on the problem, it's become apparent that symmetry requires the addition of "is not able to be" to the first sentence, a la... A removable medium (rmb) bit set to zero indicates that the medium is not removableable to be removed only if that removal is accompanied by the breaking of the I_T nexus connection between the application client to the device server. A rmb bit set to one indicates that the medium is removable able to be removed without affecting the I_T nexus between the application client and the device server. All the best, .Ralph From: John Geldman > Sent: Thursday, July 30, 2020 12:37 AM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. Ralph, I seem to be missing something. >From my world view RMB was/is about removable media, not removable devices. We still want it to be. If we declare that rmb set to zero means not able to remove the media, then to remove the media, the device server has to be removed, i.e., disconnected. The statement that removal of a device breaks the connection to the host needs to be said? Isn't the problem folks thinking RMB was/is about removable devices? John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 6:45 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition IMHO The absence of any statement about the I_T nexus in the RMB=0 case severely limits the value of the change. From: John Geldman > Sent: Wednesday, July 29, 2020 8:27 PM To: Ralph Weber >; t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. The observations are all correct, but they do obfuscate. Does this work? A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed from the device server. A rmb bit set to one indicates that the medium is removableable to be removed from the device server. If the rmb bit is set to one and the medium is removed, then the I_T nexus between the application client and the device server, if any, is not affected. John From: t10-bounces at t10.org > On Behalf Of Ralph Weber Sent: Wednesday, July 29, 2020 5:52 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I agree that the two sentence are not symmetrical, however... I cannot find a way to make them symmetrical that does not involve an excess of Rube Goldberg English that obfuscates more than it clarifies. Please feel free to augment the suggestion made below with detailed text changes. All the best, .Ralph From: Kevin D Butt > Sent: Wednesday, July 29, 2020 3:42 PM To: Ralph Weber > Cc: t10 at t10.org Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION: This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. That is starting to look good to me, but I would suggest making it symmetrical. A removable medium (rmb) bit set to zero indicates that the medium is not removable able to be removed without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 13:20 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Or... how about something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable without removing the I_T nexus that connects the application client to the device server. A rmb bit set to one indicates that the medium is removableable to be removed without affecting the I_T nexus between the application client and the device server. From:Curtis Stevens > Sent: Wednesday, July 29, 2020 3:12 PM To: Ballard, Curtis C (HPE Storage) >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition CAUTION:This email originated from outside of Western Digital. Do not click on links or open attachments unless you recognize the sender and know that the content is safe. I think part of the blatant wording that Ralph is proposing is that those making the requirements on the device side do not fully understand SCSI. This has been the case from day 1 and has not changed. It is fairly easy to understand commands and what they do. Understanding the architecture is a different story. I think your suggested change could be integrated with Ralphs change... --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:Ballard, Curtis C (HPE Storage) > Sent: Wednesday, July 29, 2020 12:49 PM To: Curtis Stevens >; Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org> Subject: RE: [T10] Revisiting the RMB (Removable Medium) bit definition Why don't we specify it in architectural terms. Something along the lines of: A removable medium (rmb) bit set to zero indicates that the medium is not removable from the device server. A rmb bit set to one indicates that the medium is removableable to be removed without removing the device server. Curtis Ballard Hewlett Packard Enterprise HPE Storage R&D Fort Collins, CO (970) 898-3013 From:t10-bounces at t10.org[mailto:t10-bounces at t10.org] On Behalf Of Curtis Stevens Sent: Wednesday, July 29, 2020 1:04 PM To: Kevin D Butt >; Ralph Weber > Cc: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would turn the proposal around and say the RMB bit shall only be set to one if the device is able to report ... after is media is ejected. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Wednesday, July 29, 2020 8:48 AM To: Ralph Weber > Cc: t10 at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, I applaud your goal, however, I am concerned about this definition. Take, for example, a tape library. It has removable media, yet it would not report one of these two additional sense codes. Even if all user cartridges were removed, the library would still be ready. It could do management type things including using diagnostic cartridges (not seen as medium to the application client) to test drives. The RMBshould be set to zero, if the device server is not able to terminate a TEST UNIT READY command (6.48) with CHECK CONDITION status, with the sense key set to NOT READY and the additional sense code: a)set to NOT READY, MEDIUM NOT PRESENT; or b)having the ADDITIONALSENSECODEfield (see 4.4.2 and 4.4.3) set to 3Ah (e.g., MEDIUM NOT PRESENT - TRAY OPEN). Thanks, Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Ralph Weber > To: "t10 at t10.org" > Date: 07/29/2020 06:49 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ Proposal 20-082r0SPC-6: Removable Medium Bit Expectations has been posted to bring this discussion thread to the attention of CAP. Resisting the ongoing calls to put as many words as possible into every standard, the proposal includes only one of the methods for defining a "real" removable medium device discussed in this thread. For the purposes discussed here, only one method is necessary to sufficiently define the standard. All the best, .Ralph From:t10-bounces at t10.org > On Behalf Of Michael Webster Sent: Tuesday, July 21, 2020 7:57 PM To: t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would like to add one other tidbit: A device reporting RMB=1 in addition to being capable of a state that reports MEDIUM NOT PRESENT, such a device should also properly support and execute an ejection operation of that media when sent a START/STOP UNIT command with Power_Condition=0, LoEj=1, & Start=0. Mike Webster Western Digital From: > on behalf of Mike Webster > Date: Friday, July 17, 2020 at 10:30 AM To: "t10 at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I would expect a USB memory stick to report RMB=0 and ejection causes the whole device to be removed from the operating system's driver stack. Afterward, such an ejected device would require an unplug/re-plug for the device to respond to anything on USB. Unless that USB memory stick included a slot for media (e.g. an SD slot) then I would expect the USB memory stick to report RMB=1 and ejection would only cause the device to remove the media, the device would remain in the operating system's driver stack, and the device would begin reporting CHECK CONDITION, NOT READY, MEDIUM NOT PRESENT for TEST UNIT READY. Mike Webster Western Digital From: > on behalf of Curtis Stevens > Date: Friday, July 17, 2020 at 10:05 AM To: Kevin D Butt >, Paul Suhler > Cc: "t10 at t10.org" >, "t10-bounces at t10.org" > Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition We had this descussion during the original specification development. For USB, the device controller is removed along with the media. There is nothing present to respond with an RMB bit. If however, you plug in a USB floppy, you can have an RMB bit that enables eject function reporting. --------------------------------------------- Curtis E. Stevens Technologist Seagate Technology E-Mail: Curtis.Stevens at Seagate.com Phone: 949-307-5050 ________________________________ From:t10-bounces at t10.org> on behalf of Kevin D Butt > Sent: Friday, July 17, 2020 9:40 AM To: Paul Suhler > Cc: t10 at t10.org>; t10-bounces at t10.org> Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition I concur with Curtis and Paul. Kevin D. Butt SCSI Architect, Tape Firmware, Data Retention Infrastructure T10 Standards MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744 Tel: 520-799-5280 Fax: 520-799-2723 (T/L:321) Email address: kdbutt at us.ibm.com =========== Interesting Links =========== [ IBM Tape Storage ] https://www.ibm.com/it-infrastructure/storage/tape [ SSIC - HBA/OS/Switch/Product interoperation ] https://www-304.ibm.com/systems/support/storage/ssic/interoperability.wss [ LTO & 3592 ISV Support Matrix ] www.ibm.com/systems/resources/lto_isv_matrix.pdf [ LTO SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003556 [ 3592 SCSI Reference ] http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003248 =================================== From: Paul Suhler > To: "Ballard, Curtis C (HPE Storage)" >, Ralph Weber >, "t10 at t10.org" > Date: 07/17/2020 09:19 Subject: [EXTERNAL] Re: [T10] Revisiting the RMB (Removable Medium) bit definition Sent by: t10-bounces at t10.org ________________________________ I agree with Curtis about the need to be able to report NOT READY, MEDIUM NOT PRESENT. I've worked on both SSC and SBC devices like that. Paul Kioxia From:t10-bounces at t10.org > On Behalf Of Ballard, Curtis C (HPE Storage) Sent: Friday, July 17, 2020 9:54 AM To: Ralph Weber >; t10 at t10.org Subject: Re: [T10] Revisiting the RMB (Removable Medium) bit definition Ralph, That bit is used with tape drives and was used with SCSI magneto optical drives, DVD drives, etc. Even system I've used with that bit set to one, and I've worked with several devices from different manufacturers that set that bit to one, only use it in reference to the media with no device server disruption other than the expected Unit Attention condition transitions as the media is removed and re-loaded. SCSI devices should only set that bit to one if they are able to report an 02h/3Ah/00h - NOT READY, MEDIUM NOT PRESENT Curtis Ballard Hewlett Packard Enterprise From:t10-bounces at t10.org> On Behalf Of Ralph Weber Sent: Thursday, July 16, 2020 5:27 PM To: t10 at t10.org Subject: [T10] Revisiting the RMB (Removable Medium) bit definition Regarding the Standard INQUIRY data RMB bit, SPC-6 r02 says... "A removable medium (RMB) bit set to zero indicates that the medium is not removable. A RMB bit set to one indicates that the medium is removable." An interesting debate has developed regarding whether a USB "memory stick" is an RMB=0 or an RMB=1 device. One wag has suggested that the RMB bit definition be conditionalized on the presence/absence of a coincident Logical Unit Reset condition or I_T Nexus Loss condition, something along the lines of... "A removable medium (RMB) bit set to zero indicates that the medium is not removable without resulting in a Logical Unit Reset condition (see SAM-6) or an I_T Nexus Loss condition (see SAM-6). A RMB bit set to one indicates that the medium is removable with no concurrent Logical Unit Reset condition or I_T Nexus Loss condition." How says the T10 body politic? Thanks, .Ralph_______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 _______________________________________________ T10 mailing list T10 at t10.org https://www.t10.org/mailman/listinfo/t10 -------------- next part -------------- An HTML attachment was scrubbed... URL: