Minutes of SSC/SMC WG Mtg, Palm Springs (96-274r0)

Erich Oetting erich_oetting at stortek.com
Mon Nov 11 10:20:17 PST 1996

* From the SCSI Reflector (scsi at symbios.com), posted by:
* erich_oetting at stortek.com (Erich Oetting)
Accredited Standards Committee          Doc. Number X3T10/96-274-r0
X3, Information Technology

To:       Membership of X3T10

From:     Erich Oetting

Subject:  Minutes of SSC/SMC Working Group Meeting
                 Palm Springs, CA, Nov. 8, 1996


1. Opening Remarks

2. Approval of Agenda

3. Attendance and Membership

4. SSC Topics

4.1 Data Transfer and ILI.
4.2 Write FM buffer immediate.
4.3 Write Append Only. (96-275-r0)
4.4 SSC Partition Change Control. (96-265-r0)
4.5 Prevent/Allow Setting. (96-266-r0)
4.6 New Mode of operation

5. SMC Topics

5.1 Read Element Status and Reservation Conflicts. (96-267-r0)
5.2 World Wide names.

6.   Other Topics.

7.    Meeting Schedule.

8.    Adjournment.

                              Results of Meeting

1.    Opening Remarks

Ted Lappin, the SSC Technical Editor, called the meeting to order at 9:00
a.m., Thursday, Nov. 8, 1996.  He thanked Norm Harris of Adaptec for
arranging and hosting the meeting.

As is customary, the people attending introduced themselves and a copy of
the attendance list was circulated.

The draft agenda was approved.

3.    Attendance and Membership

Attendance at working group meetings does not count toward minimum
attendance requirements for X3T10 membership.  Working group meetings are
open to any person or organization directly and materially affected by
X3T10's scope of work.

The following people attended the meeting:

       Name       Organization        Email Address
--------------    -------------       -------------
Edward Lappin      Exabyte           tedl at exabyte.com
Erich Oetting      Storage Tech.     erich_oetting at stortek.com
Ken Hallam         Unisys            Ken.Hallam at mv.unisys.com
Connie Kephart     Exabyte           conniek at exabyte.com
Lee Jesionowski    IBM               jez at vnet.ibm.com
Doug Hagerman      Digital           hagerman at starch.enet.dec.com
Ralph Weber        Symbios Logic     ralph_weber at symbios.com
Rob Basham         IBM               robbyb at vnet.ibm.com
Charles Monia      Digital           MONIA at shp.dec.com
David Cressman     Quantum           dcressm at tdh.qntm.com

Total of 10 People Present

4. SSC Topics

4.1 Data Transfer and ILI.

Ted asked if data should be transferred when a fixed mode READ command
encounters a record that does not match the fixed block size.  The group
agreed that the smaller of the actual record size or fixed block size
should be transferred, and then status should be returned.  Ted will
clarify the wording in SSC.

4.2 Write FM buffer immediate

Ted will add language to SSC clarifying that a Write Filemarks command
with a count of zero and the Immediate bit set does not necessarily flush
the buffer.  (The action is vendor specific.)

4.3 Write Append Only. (96-275-r0)

Erich talked about adding an append only bit to the Device Configuration
page.  The group had concerns about the increasing complexity of this page
and the exact definition of what operations would be allowed in "append
only mode".  (The current proposal allows trailing tape marks to be
overwritten.)  Discussion of this will be continued on the SCSI reflector
and at the next working group meeting.

4.4 SSC Partition Change Control. (96-265-r0)

Robert talked about his proposal to add a mode sense bit to indicate that
a format command must be issued after the mode select to actually change
the partition size.  Robert will rework to proposal to better specify
error reporting and other restrictions on using the new bit. The
possibility of expanding the number of partitions beyond 256 was also

4.5 Prevent/Allow Setting Outstanding.  (96-266-r0)

Robert's proposal would allow the host to determine when media removal is
prevented by adding mode sense bits.  The main concern is that a failing
initiator could leave the drive with media removal prevented until the
next reset. Doug H. felt that the entire command was broken, and should be
replaced instead. The proposal will be discussed on the reflector and at
the next working group meeting.

4.6 New Mode of Operation. (96-264-r0).

How to handle an "automatic" media changer that loads the next tape after
an unload was discussed. Some means of detecting that such an autoloader
is attached and if a new tape is available is desired. Erich pointed out
that an unload may also fail in some configurations when no room is
available for the unloaded tape.  Ralph suggested adding a couple of
ASC/ASCQ's to cover the error conditions may solve the problem. Discussion
will continue on the reflector.

4.7 Fibre Channel link error recovery.

David talked about error detection and recovery after link errors with
fibre channel attached tape drives.  With a class 3 connection, some
errors may only be detected by an upper level timeout.  With tape
commands, this timeout could be orders of magnitude longer than disk

A discussion was then held on how recovery from link errors could be
handled at the command layer. One option is to have the upper level driver
keep track of tape position, then issue READ POSITION after an error to
recover position and reissue commands as needed. The second was having the
drive attempt to reposition after detecting link errors. The group pointed
out problems with both options. Discussion of this issue will continue on
the reflector.

The group agreed that the real solution is to use class 2 for tape
devices. David will create a proposal for class 2 operation of tape
devices to be added to PLDA.

5. SMC Topics

5.1 Read Element Status and Reservation Conflicts. (96-267-r0)

Robert proposed that READ ELEMENT STATUS be allowed even when reserved by
another host, and that a bit be added to indicate that an element is
reserved by some host.  This would make the discovery process easier in
multiple host configurations. It was pointed out that some READ ELEMENT
STATUS commands may invoke changer movement to read Volume Tags and such.
Robert will modify his proposal to only allow benign READ ELEMENT STATUS
commands to be executed when reserved by another host.

5.2 World Wide names.

Erich discussed adding a feature to the READ ELEMENT STATUS command to
return device identification descriptors as defined in SPC along with the
Data Transfer Element and Import/Export Element data.  The current fields
for SCSI ID, LUN, etc would be made obsolete.  Erich will bring a detailed
proposal to the next meeting.

6.   Other Topics

No other topics were discussed.

7.    Meeting Schedule

The next meeting of the SSC/SMC Working Group will be next January in
Dallas, Texas.  Actual meeting time will be determined by the X3T10

8.    Adjournment

The meeting was adjourned at 12:20 am on Thursday Nov. 8, 1996.

* For SCSI Reflector information, send a message with
* 'info scsi' (no quotes) in the message body to majordomo at symbios.com

More information about the T10 mailing list