Re: [storm] I_T nexus identifier editorial

> I have just taken a quick look at Table A.6 on SAM-5 Annex A.  I think it
> has a potential issue: the verbiage restricts the hex prefix to just the
> lower case, i.e. "0x".  In contrast, RFC 3720 and the consolidated draft
> explicitly allow either upper or lower case - "0x" or "0X" hex prefix.  I
> believe the best place to fix this disconnect would be in the SAM-5 spec.

I agree - RFC 3720 is the normative reference for this format (to be superseded by
the consolidated iSCSI draft), whereas Annex A of SAM-5 is informative.  I suspect
that allowance of both upper case and lower case 'x' was missed in putting Annex A
of SAM-5 together and a corrective proposal to T10 is in order.  I'll work offline
with some other T10 participants to pull that corrective proposal together.

The current plan is to submit both the iSCSI consolidated draft and iSCSI new
features (SAM) draft to our AD and the IESG sometime in December.

Thanks,
--David

> -----Original Message-----
> From: Mallikarjun Chadalapaka [mailto:cbm@chadalapaka.com]
> Sent: Tuesday, November 22, 2011 1:49 PM
> To: Black, David; storm@ietf.org
> Subject: RE: [storm] I_T nexus identifier editorial
> 
> David,
> 
> I have just taken a quick look at Table A.6 on SAM-5 Annex A.  I think it
> has a potential issue: the verbiage restricts the hex prefix to just the
> lower case, i.e. "0x".  In contrast, RFC 3720 and the consolidated draft
> explicitly allow either upper or lower case - "0x" or "0X" hex prefix.  I
> believe the best place to fix this disconnect would be in the SAM-5 spec.
> 
> It also brings up the question of missing "0x" in the iSCSI I_T nexus
> identifier description of RFC 3720.  I can only surmise we must had
> considered it "obvious" in the past as the previous bullet in the text
> requires it for port names, and then defines that I_T nexus identifier is
> based on port names.  Regardless, It'd be good to fix this discrepancy now.
> So the new verbiage would be:
> 
> I_T nexus - a relationship between a SCSI Initiator Port and a SCSI Target
> Port, according to [SAM2]. For iSCSI, this relationship is a session,
> defined as a relationship between an iSCSI Initiator's end of the session
> (SCSI Initiator Port) and the iSCSI Target's Portal Group. The I_T nexus can
> be identified by the conjunction of the SCSI port names or by the iSCSI
> session identifier SSID. iSCSI defines the I_T nexus identifier to be the
> tuple (iSCSI Initiator Name + ",i,0x" + ISID in text format, iSCSI Target
> Name + ",t,0x" + Portal Group Tag in text format) - an upper case hex prefix
> "0X" may alternatively be used in place of "0x".
> 
> 
> As for publishing a new version, I could try to get to it soon..... (the
> main challenge is the text conversion process: it creates a couple of
> hundred broken alignments, and ASCII figure skews, most of which are caught
> by idnits and need manual fixing, :( )
> 
> Thanks.
> 
> Mallikarjun
> 
> 
> 
> -----Original Message-----
> From: david.black@emc.com [mailto:david.black@emc.com]
> Sent: Monday, November 21, 2011 4:02 PM
> To: cbm@chadalapaka.com; storm@ietf.org
> Cc: david.black@emc.com
> Subject: RE: [storm] I_T nexus identifier editorial
> 
> Mallikarjun,
> 
> I think that's enough of a technical change that making it now would be a
> good idea, especially as we can't send the consolidated draft for
> publication until the new features (SAM) draft is ready to go also.
> 
> While you're looking at this, please double check the alignment of these
> names and identifiers with the material in Annex A of SAM-5 (e.g., both the
> ISID and TPGT are documented there as hex numbers with a '0x' prefix).
> 
> Thanks,
> --David (storm WG co-chair).
> 
> > -----Original Message-----
> > From: storm-bounces@ietf.org [mailto:storm-bounces@ietf.org] On Behalf
> > Of Mallikarjun Chadalapaka
> > Sent: Monday, November 21, 2011 6:44 PM
> > To: storm@ietf.org
> > Subject: [storm] I_T nexus identifier editorial
> >
> > In responding to editorial comments on iSCSI-SAM-4 draft, I have just
> > noticed a related editorial in the consolidated draft that dates back
> > to RFC 3720.
> >
> > I_T nexus - a relationship between a SCSI Initiator Port and a SCSI
> > Target Port, according to [SAM2]. For iSCSI, this relationship is a
> > session, defined as a relationship between an iSCSI Initiator's end of
> > the session (SCSI Initiator Port) and the iSCSI Target's Portal Group.
> > The I_T nexus can be identified by the conjunction of the SCSI port
> > names or by the iSCSI session identifier SSID. iSCSI defines the I_T
> > nexus identifier to be the tuple (iSCSI Initiator Name + 'i' + ISID,
> > iSCSI Target Name + 't' + Portal Group Tag).
> >
> > The parenthetical expression at the end should really have four commas
> > in
> > it:
> >
> > (iSCSI Initiator Name + ',i,' + ISID, iSCSI Target Name + ',t,' +
> > Portal Group Tag).
> >
> > This is how the mapping table captures the nexus identifier in the
> > iSCSI-SAM-4 draft.  And it'd be good to have consistency between the
> > two drafts.  I am assuming it'd be possible to make this tweak during
> > the authors' 48 hour review before the publication (David/Tom?)
> >
> > Thanks.
> >
> > Mallikarjun
> >
> >
> > _______________________________________________
> > storm mailing list
> > storm@ietf.org
> > https://www.ietf.org/mailman/listinfo/storm
> 
> 