<div dir="ltr">Gents,<div><br></div><div>I'm sending this email for archiving purposes:</div><div>It's been a while since my last update on this topic. It turned out that although only one node, randomly, at a time lost connection to storage, the issue was not at all with ovirt, but with the storage.</div>
<div><br></div><div>I'd like to refer to these 2 topics for more information: <a href="http://lists.freebsd.org/pipermail/freebsd-net/2014-March/038061.html">http://lists.freebsd.org/pipermail/freebsd-net/2014-March/038061.html</a> and <a href="http://lists.freebsd.org/pipermail/freebsd-net/2014-February/037967.html">http://lists.freebsd.org/pipermail/freebsd-net/2014-February/037967.html</a>.</div>
<div><br></div><div>Keywords: ovirt freebsd 9.2 zfs ixgbe intel 10gbit NFS</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Feb 24, 2014 at 3:55 PM, Ronen Hod <span dir="ltr"><<a href="mailto:rhod@redhat.com" target="_blank">rhod@redhat.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">On 02/24/2014 11:48 AM, Nir Soffer wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
----- Original Message -----<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
From: "Johan Kooijman" <<a href="mailto:mail@johankooijman.com" target="_blank">mail@johankooijman.com</a>><br>
To: "Nir Soffer" <<a href="mailto:nsoffer@redhat.com" target="_blank">nsoffer@redhat.com</a>><br>
Cc: "users" <<a href="mailto:users@ovirt.org" target="_blank">users@ovirt.org</a>><br>
Sent: Monday, February 24, 2014 2:45:59 AM<br>
Subject: Re: [Users] Nodes lose storage at random<br>
<br>
Interestingly enough - same thing happened today, around the same time.<br>
Logs from this host are attached.<br>
<br>
Around 1:10 AM stuff starts to go wrong again. Same pattern - we reboot the<br>
node and the node is fine again.<br>
</blockquote>
So we made some progress, we know that it is not a problem with old kernel.<br>
<br>
In messages we see the same picture:<br>
<br>
1. sanlock fail to renew the lease<br>
2. after 80 secodns, kill vdsm<br>
3. sanlock and vdsm cannot access the storage<br>
4. kernel complain about nfs server timeouts<br>
(explains why sanlock failed to renew the lease)<br>
5. after reboot, nfs is accessible again<br>
6. after few days goto step 1<br>
<br>
This looks like kernel nfs issue.<br>
<br>
Could be also kvm issue (running bsd on the one of the vm?)<br>
<br>
Could be also some incompatibility with the nfs server - maybe you are using<br>
esoteric configuration options?<br>
<br>
CCing Ronen, in case this is related to kvm.<br>
</blockquote>
<br></div></div>
Not seems to be related to KVM.<br>
Adding Ric Wheeler.<br>
<br>
Ronen.<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
thread: <a href="http://lists.ovirt.org/pipermail/users/2014-February/021507.html" target="_blank">http://lists.ovirt.org/<u></u>pipermail/users/2014-February/<u></u>021507.html</a><br>
<br>
Nir<br>
</blockquote>
<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>Met vriendelijke groeten / With kind regards,<br>Johan Kooijman<br><br>T +31(0) 6 43 44 45 27<br>F +31(0) 162 82 00 01<br>E <a href="mailto:mail@johankooijman.com">mail@johankooijman.com</a>
</div>