Well, if you read the mailing list in the past weeks you see some people who have problems with vms on gluster. <div><br></div><div><a href="http://www.gluster.org/pipermail/gluster-users/2012-July/033752.html" target="_blank">http://www.gluster.org/pipermail/gluster-users/2012-July/033752.html</a><br>
</div><div><a href="https://bugzilla.redhat.com/show_bug.cgi?id=842254" target="_blank">https://bugzilla.redhat.com/show_bug.cgi?id=842254</a><br></div><div><a href="http://www.gluster.org/pipermail/gluster-users/2012-September/034298.html" target="_blank">http://www.gluster.org/pipermail/gluster-users/2012-September/034298.html</a><br>
</div><div><br></div><div>My experience shows that it most of the time happens with large instances (disk file > 50gb) and under I/O load.<div>
The question is, how to debug such a scenario? </div><div><br></div>
</div>
<div class="gmail_extra"><br><br><div class="gmail_quote">2012/11/27 Gerald Brandt <span dir="ltr"><<a href="mailto:gbr@majentis.com" target="_blank">gbr@majentis.com</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im"><br>
><br>
> Hi Whit, To be honest, I don't see any improvements since 3.2<br>
> concerning virtual machines (maybe under the hood?). Self-heal is<br>
> still blocking and performance is not better.<br>
> We've waited over 6 months for the 3.3 release and it was really<br>
> disappointing for me personally.<br>
><br>
><br>
> Cheers,<br>
> Christian<br>
><br>
<br>
</div>Hi,<br>
<br>
Something is seriously wrong there. I just rebuilt a RAID, with complete loss of data. GlusterFS self healed/replicated with no stutters or downtime to the 23 VM's I run on the Gluster mirror.<br>
<span class="HOEnZb"><font color="#888888"><br>
Gerald<br>
</font></span></blockquote></div><br></div>