[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: sylsog in NETCONF notifications
Rainer Gerhards wrote:
I am a bit confused. You say:
My intent is to have a content-independent framework for delivering
notifications, but not have the NETCONF WG develop any standard
notification content unless it is directly related to configuration.
I fully agree with you on that. But you also say:
IMO, syslog is important enough that it is worthwhile to have
filtering mechanisms that actually work for syslog.
Generalized filters sound great, but if they are impossible
to use for the most common application (syslog/RAW), then
they are mostly irrelevant. Telling all the vendors "Hey, I have a
really cool XML tool, so if you could re-design and re-code
15,000 or so syslog messages to output an XML subtree instead of text,
well, that would be great."
I suggest coming up with a transition plan, because it will
be about a decade before you get to use your fancy XML tools on
I agree with you on the need and the timing. I am not sure if I am
with you who should do that. Are you saying the NETCONF WG should do
that? If so, I think it is beyond the NETCONF charter.
I would appreciate if you could clarify if you think NETCONF should
a) develop a notification mechanism and a data model for NETCONF
not sure what you mean by 'data model'.
Yes -- if that means a data model for the
NV-stored configuration parameters for notification delivery,
and perhaps some state data related to notification delivery.
b) develop a data model and transformation rules for SYSLOG
If you mean transformation of all ad-hoc syslog text content
to a standard and structured XML hierarchy -- no. That is way out
of scope. However, filter parameters to suppress notification delivery
are in scope. (The non-normative appendices are supposed to be removed.)
to unsubscribe send a message to firstname.lastname@example.org with
the word 'unsubscribe' in a single line as the message text body.