Re: iSCSI 04 queries

[julian_satran@il.ibm.com]

Mark,

1-cmdSN is there in the Text command because (at least in theory) you may
want to have it reach
the target in order - and some of the key=value pair may have an active
semantic
(there is one that I missed already - and it is called SendTargets that is
be being proposed by the ND team for target discovery )
2 - Mea culpa!  It is a residue from 03 that I missed after three passes -
just ignore the b7 - the new meaning is specified above in 2.7.1 and it is
now significant as final for both input and output. THANX

3. I would say that format error is good enough (the same goes for a wrong
LUN, a wrong tag etc.).

Julo

Mark Burton <markb@ordern.com> on 24/02/2001 02:45:36

Please respond to Mark Burton <markb@ordern.com>

To:   ips@ece.cmu.edu
cc:
Subject:  iSCSI 04 queries





Hi,

I would be very grateful if anyone can supply answers to these
questions (prompted by my reading of draft-ietf-ips-iSCSI-04):

1 - Why is the cmdSN field required in the iSCSI TEXT PDU?

2 - Section 2.7.5 (SCSI Data) talks about a P bit, but that bit isn't
    shown in the PDU diagram, what's wrong?

3 - There doesn't seem to be an the appropriate error code to return
    when an initiator provides invalid values for TSID, ISID or CID?

Thanks,

Mark