Re: draft-ietf-isis-traffic

[Alex Zinin <zinin@psg.com>]

Thursday, January 09, 2003, 7:17:49 AM, Thomas Narten wrote:
>>From a quick review,

> non-derivative rights clause needs discussion.

> Remind me why the WG is doing what IANA should be doing...

from the discussion on the call, we agreed that I would
ask the authors to put words in text asking IANA to
maintain an info-only registry. Now I'm thinking, since
IANA will maintain the registry and the ISIS WG will do
the allocation, the registry may as well be regular,
with the ISIS WG performing review of the allocation
requests. Since this is IP-specific, we shouldn't have
objections from JTC1, though I'll check.

> nit: 2119 ref and definitions pointer needs to be in the document
> proper (it's in the boilerplate)

will ask

> ipv6 is not included. Should it?

Nope, IPv6 is addressed in another document.

>>    This sub-TLV is OPTIONAL. This sub-TLV SHOULD appear at most once in
>>    each extended IS reachability TLV.

> There are  number of sub-options  that say "SHOULD appear at most
> once". But it is not a protocol violation to send more than one (since
> it is not MUST). What is the correct behavior for the receiver?
> Specifying this would seem to be necessary to get good
> interoperability predictability

I asked them this question. WIll check again.

>> Security Considerations
>> 
>>    This document raises no new security issues for IS-IS.

> Oh? None of the new information that is included adds anything new?

I'll ask the authors to add some words on increased LSP
origination frequency due to new type of info announced
and how to make sure this doesn't overload the IS-IS
domain.

>>            250-254                     Reserved for cisco specific extensions

> Excuse me?

I will ask the authors to check with Cisco if they are
using this region at all, as I mentioned on the call.


Alex