Re: iSCSI - working draft and IANA

["John Hufferd" <hufferd@us.ibm.com>]

Mark,
Please explain about the need to register the current ones, what will that
buy us?

.
.
.
John L. Hufferd
Senior Technical Staff Member (STSM)
IBM/SSG San Jose Ca
Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688
Home Office (408) 997-6136, Cell: (408) 499-9702
Internet address: hufferd@us.ibm.com


Mark Bakke <mbakke@cisco.com>@ece.cmu.edu on 07/30/2002 10:52:20 AM

Sent by:    owner-ips@ece.cmu.edu


To:    Julian Satran/Haifa/IBM@IBMIL
cc:    Steve Senum <ssenum@cisco.com>, ips@ece.cmu.edu
Subject:    Re: iSCSI - working draft and IANA



Julian-

I'm not sure I see the need for registering keys, but that
aside, if we register digest and auth methods I would suggest
that we also register an integer method number with each; this
will make it easier to add them to the MIBs.  I would also
suggest that we register the current ones as well as any
extensions.

--
Mark

Julian Satran wrote:
>
> Registering the current keys is an issue already raised by Mallikarjun.
The only think I can
> comment about is that I don't see what we stand to gain (and I can
cleraly see the  pain!).
>
> As for the prefixes - they are aimed at clearly delineating what is
mandatory (key defined in the
> basic iSCSI doc) from vendor or group=of-vendors additions.
>
> The registration is meant to allow groups of vendors to agree on a key
and provide a semantic doc
> (an RFC that can be informational).
>
> Julo
>
>   Steve Senum <ssenum@cisco.com>
>                                            To:        Julian
Satran/Haifa/IBM@IBMIL
>   07/30/2002 01:48 AM                      cc:        ips@ece.cmu.edu
>                                            Subject:        Re: iSCSI -
working draft and IANA
>
>
>
> Julian,
>
> 1. I would suggest registering all the current iSCSI keys, auth methods,
> and digests with the IANA, with references to the iSCSI RFC (when
published),
> and dropping the X#, Y#, and Z# prefixes.  This would be more consistent
> with how I have seen this done in the past.
>
> 2. I am not sure I really see the need.  In other cases, this is done
> to allow vendor specific registrations, but we already have a mechanism
> for that (the X- prefix).  Note that there is no reason why a vendor
> can't defined a vendor specific key in an informational RFC.
>
> Regards,
> Steve Senum
>
> Julian Satran wrote:
> >
> > Dear colleagues,
> >
> > The current (today's) version of the draft has a revised IANA
consideration
> > section
> > and specific  indication on how to build keys, authentication methods
and
> > digests.
> >
> > David Black suggested that we might want to go for 3 different
registries
> > maintained by IANA for iSCSI
> > and I liked the idea.
> >
> > Please comment,
> > Julo

--
Mark A. Bakke
Cisco Systems
mbakke@cisco.com
763.398.1054