[Gluster-infra] Side effect of the metldown fix for regression

Michael Scherer mscherer at redhat.com
Wed Jan 10 17:18:27 UTC 2018


Hi,

so it seems that the meltdown fix did slowdown regression testing, as
told to me by Nigel (after Nithya pointed this looked like timing
issue). 

After a quick discussion on irc, we suspect this might be the lack of
pcid for the rackspace hosts.

So the good news is that we have a fix:
 https://twitter.com/berrange/status/950209752486817792


The bad news is that we can only add it to builder in the cage, which
is unfortunate.

A ticket was opened to rackspace by Nigel, and I will watch over it to
have it fixed ASAP.  

In the mean time, I am taking care of fixing it in the cage as a backup
plan, but that might need a reboot of the various builders (and that's
always fun to hit "put that node offline" 40 times 2 times, because
jenkins is hard to automate with our configuration).

-- 
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://lists.gluster.org/pipermail/gluster-infra/attachments/20180110/2e8acbe0/attachment.sig>


More information about the Gluster-infra mailing list