[Gluster-devel] [Gluster-infra] Unexpected jenkins restart
mscherer at redhat.com
Tue May 26 11:04:27 UTC 2020
Le mardi 26 mai 2020 à 12:50 +0200, Michael Scherer a écrit :
> Le mardi 26 mai 2020 à 12:33 +0200, Michael Scherer a écrit :
> > Hi,
> > while working on the jenkins automation (mostly adding new host
> > with
> > CLI), I did a jenkins-cli reload-configuration. Turn out that this
> > is,
> > contrary to what I expected, also seems to have blocked jenkins
> > somehow.
> > I am trying to figure exactly what is happening, cause I do not
> > expect
> > that to be so long.
> My current hypothesis is that jenkins is still running, but currently
> reloading the config seems to imply "read all xml files in
> /var/lib/jenkins" or something. I do not know where this do come
> but this also mean that the server is busy parsing all builds results
> in /var/lib/jenkins/jobs, a 72G directory with around 148652 files if
> can trust find and wc.
> I will dig a bit more before interrupting the server brutally
Seems jenkins did manage to come back by itself, after 20 minutes.
The good news is that it seems that after 3 days of work, we found a
way (minus some ansible hack to be fixed) to dynamically add builders.
The bad new is that jenkins reload is busted, cause 20 minutes of
downtime do not seems optimal at all, and we would need to fix that
before going on.
For folks who are eager on reading java, that's somewhere here:
( from ./core/src/main/java/hudson/cli/ReloadConfigurationCommand.java
to doReload, to reload to loadTasks).
Michael Scherer / He/Il/Er/Él
Sysadmin, Community Infrastructure
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 836 bytes
Desc: This is a digitally signed message part
More information about the Gluster-devel