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

[Marco Marzetti <[e-mail]>]

Robert,

Despite i like the idea to use this feature in automation, it sounds
overkill to me to force TLVs.
Wouldn't be simpler to define the message as UTF8 string and let
operators do what they think is more reasonable?

AFAIK we don't have standards that defines "assurance messages" (do
we?) and this draft is not the right place to define it

Regards

On Sat, Nov 19, 2016 at 2:22 PM, Robert Raszuk <[e-mail]> wrote:
>> I'm certainly not against allowing people to send a URL
>> or other data, but our use case is likely to be the ticket# and/or
>> a phone number for contacting our NOC to streamline communication.
>
>
> Actually what drove my suggestion for URL was as usual desire for
> automation.
>
> As 128 octets of unstructured free form text is proposed that still requires
> human reading the text which is something we should try to avoid IMHO. Worse
> it requires someone to pick up the phone and make a call ...
>
> At min perhaps draft should define few optional prefixes/key words which
> could be machine recognized: "url:", "mail:", "phone:", "description:",
> "eta:", "ticket:", "duration:" etc ....
>
> Reason ... if the outage is to last 5-10 min in the middle of the night (ad
> delta from current time vs eta or duration) the analyzing script will not
> raise the alarm and will save few folks to being woken up. Simple and useful
> IMHO. Not everyone has 24/7 NOC.
>
> And btw - while good idea - adding it show bgp neighbor will likely break
> many parsers and will require yang model extension :).
>
> Cheers,
> R.
>
> PS. Do we need to extend BMP too ?
>
>
> _______________________________________________
> Idr mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/idr
>



-- 
Marco