----- Original Message -----
From: "Michael Scherer" <mscherer(a)redhat.com>
To: infra(a)ovirt.org
Sent: Wednesday, May 21, 2014 3:17:49 AM
Subject:
Lists.ovirt.org is full ( or almost )
Hi,
I just started to receive mail from postfix about insufficient storage
on
lists.ovirt.org. So this would have blocked all ML.
I did some cleaning in urgency ( yum cache, etc ), but we would still
need to remove urgently lots of space to avoid seeing the issue again
next week.
So I would like to propose the following :
- short term
- remove the iso in /home/obasan , that's ovirt 3.3, we are at 3.4, so
maybe not needed. that's 1G.
dcaro also found 9GB of archived engine-patches that we probably don't need.
deleting old releases, which are not supported anymore (3.2 and lower?)
- mid term
- do backup somewhere else than on this server. If the server crash,
having backup on it will be useless :) ( ie, the 9G of gerritt backup )
- have smarter backups. Doing a full backup of gerrit mean there is
lots of duplication. Not sure if we should backup tmp file, for example.
- remove texlive. it take half a gigabit, only for one version of
mediawiki. Which is weird on this server since it is not installed, IIRC
+1 of revisiting the backups and possbibly backup to another server (maybe alterway or
rackspace for now)?
- longer term
- move to a bigger server
- have monitoring of the root partition usage so this can be detected
before it become a issue
- try to have less automated mails sent on ml, as they take space
(logwatch for one, but maybe others)
that's on the plan to as part of migration to PHX2 lab, which will have 16TB of disk
space.
hopefully within a few weeks if all goes well.
Any opinions ?
--
Michael Scherer
Open Source and Standards, Sysadmin
_______________________________________________
Infra mailing list
Infra(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/infra