[cgl_discussion] POC Meeting Agenda 9:30 PST

Venkata Jagana jagana at us.ibm.com
Wed Nov 5 09:24:53 PST 2003


Few comments and my recommendation on the IPv6 MIBs:


>What I was trying to address are two things:
>
>1) RFC 2452, RFC 2454, RFC 2466 will all be obsolete soon.  And my
>   work will be based on the upcoming Internet Draft, which merges
>   the IPv4 and IPv6 mib.

    It will be quite a while before they become obsolete because the
    the current internet draft is still a work-in-progress document.
    As I said before we can not reference the internet drafts as they
    are most likely to change until they are accepted as proposed
standards.

>
>2) There is no new Internet Draft for RFC 2454 (UDP mib) defined yet.
>   So I think we should down-grade it to Priority 2.
>
>I suggest we change this requirement for CGL 2.0 into something like
>the following:
>
>SVC.1.4
>Description: CGL shall provide required support for all IPv6 MIBs as
>specified by IETF IPv6 WG. The support has to be based on the new IP
>MIB standard, including:
>
>* "Management Information Base for the Internet Protocol (IP)"
>http://www.ietf.org/internet-drafts/draft-ietf-ipv6-rfc2011-update-04.tx
>t
>
>* "Management Information Base for the Transmission Control Protocol
>(TCP)"
>http://www.ietf.org/internet-drafts/draft-ietf-ipv6-rfc2012-update-04.tx
>t


My recommendation at this point is as follows:

We can not modify the existing requirement SVC 1.4 since the v2.0 reqts
document is already out.

Keep the existing requirement as it is for now but add these internet
drafts
in errata as the ones replacing the original requirement SVC 1.4 once the
drafts
become proposed standard. We should highlight the fact that the
implementation
must conform to the proposed standards of these internet drafts.

>And also we should add another requirement to specifically address the
>UDP mib issue.

The draft on UDP is just out. Don't have a ptr now but will provide soon.
Same above reasoning would apply even in this case.

Thanks,
Venkat






More information about the cgl_discussion mailing list