[T11.3] FW: Possible error in FCP3r03c document

Bob.Nixon at emulex.com Bob.Nixon at emulex.com
Fri Sep 9 09:08:00 PDT 2005


* From the T10 Reflector (t10 at t10.org), posted by:
* Bob.Nixon at Emulex.Com
*
I agree with the general direction of the discussion in this email thread, but I need to object to one of the specific recommendations:

"
FCP-4 should drop the codes from the table to avoid the problems caused by redefining the same thing twice with two different values.
"

The numeric values for FCP_RJT codes are necessary in FCP-x.  SRR and FCP_RJT are FC-4 Link Services, so the codes used for Fibre Channel Extended Link Services are not formally applicable. Apart from Routing Control (R_CTL) values and a few rules for Exchange and Sequence management specified in FC-LS clause 5, each FC-4 is required to fully specify its own FC-4 Link Services.

As suggested in this thread, it is reasonable to use the ELS Reject codes as guidance in the choice of FC-4 Reject codes, but the content of FC-4 ELSs, including code values, is required to be specified by the FC-4. 

   - bob

-----Original Message-----
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org]On Behalf Of Elliott,
Robert (Server Storage)
Sent: Thursday, September 08, 2005 4:44 PM
To: t10 at t10.org; t11_3 at mail.t11.org
Subject: RE: [T11.3] FW: Possible error in FCP3r03c document


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Server Storage)" <Elliott at hp.com>
*
This appears to be an error introduced during FCP-2 letter ballot
comment resolution.

* FCP-1 and FCP-2 through revision 4a did not define the encoding, they
just used the name "Invalid OX_ID-RX_ID Combination".
* FCP-2 revision 4b added the Link Service Reject reason code
explanation table with the incorrect encoding of 03h.
* Subsequent versions of FCP-2 and FCP-3 continued to include that
table.

The resolution of this comment (see 00-300r3) appears to be the source
of the problem:
"1.32 Crossroads 32 (E): 8.1 Pars 2, 3 (p.31)

Par 2 refers to reason code hex '09', par 3 to reason code 00092A00h.
These should be made consistent (and be either reason code & explanation
or reason code).

Response: Accepted. They shall both use reason code and explanation. The
explanation for the case defined in the second paragraph will be: 17h,
Invalid OX_ID-RX_ID combination.

Installation: After considerable study, the T10 convention of using the
names of the reason code and reason code explanation was used. The
explanation cod "invalid OX_ID-RX_ID combination" is 03h in FC-FS, so
that value was selected. Since the data was fully contained in the
reject definitions, the text and table 18 were removed.

Solution approved, October 30, 2000."

I suspect the editor saw the table in FC-FS for BA_RJT (Basic Reject),
where a reason code explanation of 03h means Invalid OX_ID-RX-ID
Combination. This is used to reply to ABTS.

For LS_RJT (Link Service Reject), a reason code explanation of 17h means
Invalid OX_ID-RX_ID Combination. 03h means "Service Parm error -
Initiator Ctl". This is used to reply to ABTX, RES, REC, RRQ, RSI, and
RSS.

FCP-4 should drop the codes from the table to avoid the problems caused
by redefining the same thing twice with two different values.

-- 
Rob Elliott, elliott at hp.com 
Hewlett-Packard Industry Standard Server Storage Advanced Technology 
https://ecardfile.com/id/RobElliott 

________________________________

	From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of
Kevin D Butt
	Sent: Tuesday, September 06, 2005 2:25 PM
	To: David Peterson
	Cc: owner-t10 at t10.org; t10 at t10.org; t11_3 at mail.t11.org
	Subject: RE: [T11.3] FW: Possible error in FCP3r03c document

	Yes.  17h is what I want. 
	
	Kevin D. Butt
	Fibre Channel & SCSI Architect, IBM Tape Firmware, 
	6TYA, 9000 S. Rita Rd., Tucson, AZ  85744
	Tie-line 321; Office: 520-799-5280, Lab: 799-5751, Fax:
799-4138, Email: kdbutt at us.ibm.com 
	
	"David Peterson" <David.Peterson at mcdata.com> 
	Sent by: owner-t10 at t10.org 

	09/06/2005 09:03 AM 
		To Kevin D Butt/Tucson/IBM at IBMUS 
		cc <t10 at t10.org>, <t11_3 at mail.t11.org> 
		Subject
		RE: [T11.3] FW: Possible error in FCP3r03c document

	I fat-fingered, 03h is what is currently specified in FCP-2/3.
As such the query should be: 
	  
	Please repond if you object to changing "Invalid OX_ID-RX_ID
combination" to 17h (i.e. not 03h) for FCP FC-4 Link Services, along
with your reasoning. 
	  
	So Kevin, I believe you are indicating a change from 03h to 17h
is ok for you, since you already use 17h. 
	  
	Sorry for any confusion... 
	  
	...Dave 
________________________________

	From: Kevin D Butt [mailto:kdbutt at us.ibm.com] 
	Sent: Tuesday, September 06, 2005 10:41 AM
	To: David Peterson
	Cc: t10 at t10.org; t11_3 at mail.t11.org
	Subject: Re: [T11.3] FW: Possible error in FCP3r03c document
	
	Dave, 
	
	I disagree with changing this.   
	
	My reasoning is that all IBM Tape drives use 17h and not 03h.
Also, 03h is already defined for Extended Link Services and while we do
not use it, some implementation out there might.  I understand that SRR
is an FC-4 Link Service and is neither a Basic Link Service nor an
Extended Link Service.  However, it is much closer to an Extrended Link
service in it's handling than it is a Basic Link Service.  This would
force a code change, and while that change would not be difficult it
would cause differences in behavior and expectations.  How will the
HBA's react to a change in the value of the response they get - and
differences depending on the device and code level?  Will this cause a
change in their behavior? 
	
	Thanks, 
	
	Kevin D. Butt
	Fibre Channel & SCSI Architect, IBM Tape Firmware, 
	6TYA, 9000 S. Rita Rd., Tucson, AZ  85744
	Tie-line 321; Office: 520-799-5280, Lab: 799-5751, Fax:
799-4138, Email: kdbutt at us.ibm.com 
	
	
	"David Peterson" <David.Peterson at mcdata.com> 
	Sent by: t11_3-bounce at mailman.listserve.com 

	09/06/2005 07:48 AM 
	
		Please respond to
	David.Peterson

		
	
		To
		<t10 at t10.org> 
		cc
		<t11_3 at mail.t11.org> 
		Subject
		[T11.3] FW: Possible error in FCP3r03c document


		


	
	
	
	Howdy, 
	 
	Below is an email that was discussed at the last FCP-3 working
group. The belief of the working group was that their implementations
used 17h for an "Invalid OX_ID-RX_ID combination" for the SRR FC-4 Link
Service (i.e., the same as for Extended Link Services). 
	 
	Also note that FC-FS/FC-FS-2 specifies a 03h for "Invalid
OX_ID-RX_ID combination" for a BA_RJT reason code explanation. 
	 
	Please repond if you object to changing "Invalid OX_ID-RX_ID
combination" to 03h for FCP FC-4 Link Services, along with your
reasoning. 
	 
	Thanks...Dave 
	
	
________________________________

	Subject: Possible error in FCP3r03c document
	
	Dave, 
	 
	I think I caught an error in the fcp3r03c document. 
	 
	In Table 17: 
	
	 
	The Reason Code explanation for "Invalid OX_ID-RX_ID
combination" is listed as a 03h. 
	However FC-LS lists this Reason Code explanation as "03h Service
Parm error - Initiator Ctl" and "Invalid OX_ID-RX_ID combination" is
listed as a 17h. 
	 
	Am I missing something? 
	 
	 
	Thanks 

	

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org





More information about the T10 mailing list