Re: [Idr] [GROW] draft-snijders-idr-shutdown-00: Drop a line in the peer's syslog at shutdown

[Job Snijders <[e-mail]>]

On Sat, Nov 19, 2016 at 03:57:06PM +0100, Robert Raszuk wrote:
> I am not going to argue with few of you here who want free form text
> and do not accept any suggestions.

No, we are just selective in which suggestions we accept. Suggestions
falling short of the mark are countered and discarded.

> But if you can not parse few defined keywords in any free form text to
> enable better automated interpretation of the message I think there is
> more problems here.

This proposal aims to accomplish a simple thing, which is to enrich an
event with additional contextual information and assist in triaging
("This is unexpected, is this a problem?" "Should we embark on a
specific procedure?") in an operator-centric workflow. By keeping it
free form, it allows enough space to experiment with structured
inter-domain signaling, and if operators find this useful a dictionary
of messages can be offered at a later date as an extension based on
operational experience.

It is absolutely is not aiming to act as a generic inter-domain
signaling protocol for operations orchestration.

> TLV within current NOTIFICATION MSG is an overkill so do not count of
> new proposal from me on it. 

Hence this is not TLV, this is LV, already explained here:
https://www.ietf.org/mail-archive/web/idr/current/msg17127.html

> Operational message solved that already.

Operational message expired in September 2012, and as such I do not
consider it an active working group document anymore.

The Operational message does not provide a strong correlation between a
Cease event and some type of additional meta-data. Same reason why
'advisory' and 'shutdown' do not preclude each other.

I'd like to ask the chairs to move 'draft-ietf-idr-operational-message-00' to 'Dead' status.

> Btw how are you going to handle it in IX ? I assume RS peers will not
> get any info if one of the parties they used to get routes from goes
> down and sends free form msg to RS ? Routes will be get withdrawn and
> that's it.

Correct. This is the expected behaviour.

Kind regards,

Job