[OPSAWG] ops-dir review: draft-ietf-opsawg-syslog-alarm-02

[Pekka Savola <[e-mail]>]

This is a solicited ops-dir review of 
draft-ietf-opsawg-syslog-alarm-02 which is currently in IETF LC.

This could be useful.  I'll describe our related usage, maybe this will help
in placing this in an operational context. For example, we process
syslog alarms just by translating them to syslog locally (snmp server
behaviour).  This output is not very useful and in order to see real-time
status of alarms and their details, we need to log on to the equipment and
screen-scrape information.  (If I'd have to guess, I'd say the vendor just
doesn't implement SNMP polling of detailed alarm information but maybe it
doesn't even exist.)

The spec appears to be simple and fine; a few editorial modifications
(below) could improve it slightly.

editorial
---------

In S 3, the term "SD ID" is first used.  It should be expanded/explained.
It seems to come from RFC5424.


In S 3 s/SD-PARARMS/SD-PARAMS/

In S 3.6, URIs must be defined by RFC1786 and its updates; that RFC has been
made obsolete; I suggest replacing that with RFC3986 which should include
what you need.

In S 6, likewise as tehe first comment, you say "IANA is requested to
register the SD-IDs", better: "IANA is requested to register the syslog
Structured Data ID Values".  Makes it easier to find the registry in the
IANA web page as well :-)