[Gluster-infra] Build.g.o VM crash, had to restart it

Michael Scherer mscherer at redhat.com
Sat Sep 17 15:40:45 UTC 2016


Hi,

so kaleb pinged me around 15h05 UTC about jenkins not responding. After
trying to slavage the VM, I did had to restart it. It was not answering
to ping, no ssh, couldn't see anything on the text console with virsh
and virt-manager. And strace on qemu did show a loop over poll,who do
not bring much clarity on the issue.

So by 15h20, I restarted jenkins, then checking all was ok, it was not.

Problem is jenkis seems to not be able to restart by itself.

We have a issue with the dns not being set when the VM start (I do push
investigation for this after the migration, cause that's network
related, so maybe a different network will work better), and usually,
jenkins need to restart after network is restarted, but not this time.

I am still searching, I will send a email when I will have found the
problem.

-- 
Michael Scherer
Sysadmin, Community Infrastructure and Platform, OSAS


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part
URL: <http://www.gluster.org/pipermail/gluster-infra/attachments/20160917/28658ba1/attachment.sig>


More information about the Gluster-infra mailing list