datatracker.ietf.org
Sign in
Version 5.3.0, 2014-04-12
Report a bug

Purge Originator Identification TLV for IS-IS
draft-ietf-isis-purge-tlv-05

Note: This ballot was opened for revision 05 and is now closed.

Summary: Needs 8 more YES or NO OBJECTION positions to pass.

Jari Arkko

Comment (2011-02-17)

Some comments from Ari Keränen who helped me in some of my reviews today:

Abstract

This document updates 3 RFCs, should say that in the abstract. Also LSP acronym
should be expanded.

[Dan Romascanu]

Comment (2011-02-14)

>   This document requests that IANA assign code point 13 for the 'Purge
   Originator Identification' TLV from the IS-IS 'TLV Codepoints
   Registry'.  The additional values for this TLV should be: IIH:n,
   LSP:y, SNP:n, Purge:y.

I assume that this is OK with this document but usually an Internet-Draft
cannot / should not make a request for a specific value, but just require a
code point value and recommend that this value be 13.

[Ralph Droms]

Comment (2011-02-16)

Balloting no objection assuming there is an easy answer to my question
(posted to draft-ietf-isis-reg-purge-00) about the IANA ISIS-TLV
registry and notifying ISO JTC1.

[Sean Turner]

Comment (2011-02-16)

Please add a pointer for where I can find the definition of system ID.

[Tim Polk]

Comment (2011-02-16)

In the Intro, this document states:

   Field experience has observed several circumstances where an IS can
   improperly generate a purge.  These are all due to implementation
   deficiencies or implementations that predate [ISO TC1] and generate a
   purge when they receive a corrupted LSP.

In the security considerations, it is noted that:

   Therefore, all implementations in a domain implementing
   authentication MUST be upgraded to receive the POI TLV before any IS
   is allowed to generate a purge with the POI TLV.

Since you need to touch every ISIS implementation in the domain anyway, is it
worth stating that implementations SHOULD be updated for consistency with [ISO
TC1] (i.e., and not generate a purge when they receive a corrupted LSP) at the
same time?