https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=36954

--- Comment #5 from Marcel de Rooy <[email protected]> ---
(In reply to David Nind from comment #4)
> I had a go at testing, not that I understand SIP and the discussion (I'm
> also not sure what "..that logging is still done as expected" looks like).

Cool. Thanks.


> What I noticed:
> 1. For most of the entries that were previously in sip-output.log, in the
> new log they have "[WARN] [undef]@[undef]", previously they didn't have
> anything.

This is related to Net::Server warns (as started by SIP2). The Trapper modules
catches those and sends them to Koha::Logger (but without the usual context
explaining the undef's).

> Comment:
> 1. For the logs for other services, there tends to be a full log, and then
> errors separated out into a separate [service]-error.log.
> 2. Should we be doing the same here, so as not to break this pattern?

Good point. At first sight we seem to follow a pattern. But looking closer
shows that it is more like a wished pattern.
The -error log should contain all daemon messages with starts, stops, respawns
etc. The -output log should contain the output from the daemonized (client)
process.

Actually, koha-indexer and koha-worker contain daemon messages in -error. But
koha-sip and e.g. koha-zebra dont. This is related to the fact that the latter
do not daemonize a simple perl process but complexer processes that fork
themselves etc. This results in an empty -error log.

So yes, I will revert a bit to the two logfiles approach but it does not really
work since it needs much more digging in the code than anticipated here.

-- 
You are receiving this mail because:
You are watching all bug changes.
_______________________________________________
Koha-bugs mailing list
[email protected]
https://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-bugs
website : http://www.koha-community.org/
git : http://git.koha-community.org/
bugs : http://bugs.koha-community.org/

Reply via email to