Phone # for Next FCL Error SSWG - 6/24/97 (really)

Jim Mcgrath jmcgrath at QNTM.COM
Fri Jun 20 17:37:18 PDT 1997


* From the SCSI Reflector (scsi at symbios.com), posted by:
* jmcgrath at qntm.com (Jim Mcgrath)
*
     As a couple of you have already pointed out, I spaced on
     the date for the SSWG call.  This is the corrected notice.

     
     
        Date: Tuesday, June 24 (not the 30)
        Time: 9 am to 11 am PST
        Phone: 608-250-8227
        Code: 496804
     
     
---------------------------------
     
     
     The following was sent previously:
     
     
     Although the next scheduled FC Error SSWG was announced earlier 
     for Tuesday, June 17, this has been postponed to Tuesday, June 
     24.  Time is 9 am to 11 am PST.  This was done o allow people 
     to gather thoughts and information on the sequence level
     error recovery discussion we had last week at the working 
     group.
     
     While we will have a new proposal revision for the meeting, the 
     focus in on getting input from host, host adaptor, and drive 
     vendors on likes and dislikes of the current proposals.
     
     Although labeled class 2 vs class 3, the real issues are 
     different:
     
     The overhead of setting up for timeout detection on every 
     IO (simple but expensive for an adaptor) or only every
     IO needing error recovery (e.g. tape) (complex, since the 
     adaptor does not know this today, so needs an interface to 
     the UP software, but less expensive to implement).  Note 
     that the problem of a disk and tape drive behind one address 
     makes this a common problem for both proposals.
     
     The extendability of this to other device types (like disk). 
     Here we need input on the system requirements to see if any 
     approach thought of so far is even feasible.
     
     The extendability of this to other protocols - this would 
     make us use FC only mechanisms, not FCP mechanisms.
     
     Problems implementing protocols on today's silicon and 
     software, and the time and work required to overcome 
     these problems.
     
     Other issues I am sure.  Let us raise our heads out of 
     the details of the proposals are set some goals and 
     constraints - that way we can move forward more quickly.
     
     Jim
     
     
*
* 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