<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    where do I find the recovery files?<br>
    <br>
    [root@ovirt1 test vdsm]# pwd<br>
    /var/lib/vdsm<br>
    [root@ovirt1 test vdsm]# ls -la<br>
    total 16<br>
    drwxr-xr-x   6 vdsm kvm    100 Mar 17 16:33 .<br>
    drwxr-xr-x. 45 root root  4096 Apr 29 12:01 ..<br>
    -rw-r--r--   1 vdsm kvm  10170 Jan 19 05:04 bonding-defaults.json<br>
    drwxr-xr-x   2 vdsm root     6 Apr 19 11:34 netconfback<br>
    drwxr-xr-x   3 vdsm kvm     54 Apr 19 11:35 persistence<br>
    drwxr-x---.  2 vdsm kvm      6 Mar 17 16:33 transient<br>
    drwxr-xr-x   2 vdsm kvm     40 Mar 17 16:33 upgrade<br>
    [root@ovirt1 test vdsm]# locate recovery<br>
    /opt/hp/hpdiags/en/tcstorage.ldinterimrecovery.htm<br>
    /opt/hp/hpdiags/en/tcstorage.ldrecoveryready.htm<br>
    /usr/share/doc/postgresql-9.2.15/html/archive-recovery-settings.html<br>
    /usr/share/doc/postgresql-9.2.15/html/recovery-config.html<br>
    /usr/share/doc/postgresql-9.2.15/html/recovery-target-settings.html<br>
    /usr/share/pgsql/recovery.conf.sample<br>
    /var/lib/nfs/v4recovery<br>
    <br>
    <br>
    [root@ovirt1 test vdsm]# locate 757a5  (disk id)<br>
/ovirt-store/nfs1/7e566f55-e060-47b7-bfa4-ac3c48d70dda/images/757a5e69-a791-4391-9d7d-9516bf7f2118<br>
/ovirt-store/nfs1/7e566f55-e060-47b7-bfa4-ac3c48d70dda/images/757a5e69-a791-4391-9d7d-9516bf7f2118/211581dc-fa98-41be-a0b9-ace236149bc2<br>
/ovirt-store/nfs1/7e566f55-e060-47b7-bfa4-ac3c48d70dda/images/757a5e69-a791-4391-9d7d-9516bf7f2118/211581dc-fa98-41be-a0b9-ace236149bc2.lease<br>
/ovirt-store/nfs1/7e566f55-e060-47b7-bfa4-ac3c48d70dda/images/757a5e69-a791-4391-9d7d-9516bf7f2118/211581dc-fa98-41be-a0b9-ace236149bc2.meta<br>
    [root@ovirt1 test vdsm]# locate 5bfb140 (vm id)<br>
/var/lib/libvirt/qemu/channels/5bfb140a-a971-4c9c-82c6-277929eb45d4.com.redhat.rhevm.vdsm<br>
/var/lib/libvirt/qemu/channels/5bfb140a-a971-4c9c-82c6-277929eb45d4.org.qemu.guest_agent.0<br>
    <br>
    <br>
    <br>
    <div class="moz-cite-prefix">On 4/29/16 10:02 AM, Michal Skrivanek
      wrote:<br>
    </div>
    <blockquote
      cite="mid:034AC28F-A06A-43B4-95E3-BB4EAD615E04@redhat.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      <div><br>
      </div>
      <div><br>
        On 29 Apr 2016, at 18:26, Bill James &lt;<a
          moz-do-not-send="true" href="mailto:bill.james@j2.com"><a class="moz-txt-link-abbreviated" href="mailto:bill.james@j2.com">bill.james@j2.com</a></a>&gt;
        wrote:<br>
        <br>
      </div>
      <blockquote type="cite">
        <div> yes they are still saying "paused" state.<br>
          No, bouncing libvirt didn't help.<br>
        </div>
      </blockquote>
      <div><br>
      </div>
      Then my suspicion of vm recovery gets closer to a certainty:)
      <div>Can you get one of the paused vm's .recovery file from
        /var/lib/vdsm and check it says Paused there? It's worth a shot
        to try to remove that file and restart vdsm, then check logs and
        that vm status...it should recover "good enough" from libvirt
        only. </div>
      <div>Try it with one first<br>
        <div>
          <div><br>
            <blockquote type="cite">
              <div> I noticed the errors about the ISO domain. Didn't
                think that was related.<br>
                I have been migrating a lot of VMs to ovirt lately, and
                recently added another node.<br>
                Also had some problems with /etc/exports for a while,
                but I think those issues are all resolved.<br>
                <br>
                <br>
                Last "unresponsive" message in vdsm.log was:<br>
                <br>
                vdsm.log.49.xz:jsonrpc.Executor/0::WARNING::<b>2016-04-21</b>
                11:00:54,703::vm::5067::virt.vm::(_setUnresponsiveIfTimeout)

                vmId=`b6a13808-9552-401b-840b-4f7022e8293d`::monitor
                become unresponsive (command timeout, age=310323.97)<br>
                vdsm.log.49.xz:jsonrpc.Executor/0::WARNING::2016-04-21
                11:00:54,703::vm::5067::virt.vm::(_setUnresponsiveIfTimeout)
                vmId=`5bfb140a-a971-4c9c-82c6-277929eb45d4`::monitor
                become unresponsive (command timeout, age=310323.97)<br>
                <br>
                <br>
                <br>
                Thanks.<br>
                <br>
                <br>
                <br>
                <div class="moz-cite-prefix">On 4/29/16 1:40 AM, Michal
                  Skrivanek wrote:<br>
                </div>
                <blockquote
                  cite="mid:656BFC5C-A6F5-4332-90AC-C039D4E9170E@redhat.com"
                  type="cite"> <br class="">
                  <div>
                    <blockquote type="cite" class="">
                      <div class="">On 28 Apr 2016, at 19:40, Bill James
                        &lt;<a moz-do-not-send="true"
                          class="moz-txt-link-abbreviated"
                          href="mailto:bill.james@j2.com">bill.james@j2.com</a>&gt;
                        wrote:</div>
                      <br class="Apple-interchange-newline">
                      <div class="">
                        <div bgcolor="#FFFFFF" text="#000000" class="">
                          thank you for response.<br class="">
                          I bold-ed the ones that are listed as
                          "paused".<br class="">
                          <br class="">
                          <br class="">
                          [root@ovirt1 test vdsm]# virsh -r list --all<br
                            class="">
                          Â Id   
                          Name                          
                          State<br class="">
----------------------------------------------------<br class="">
                        </div>
                      </div>
                    </blockquote>
                  </div>
                  <div><br class="">
                  </div>
                  <div>
                    <blockquote type="cite" class="">
                      <div class="">
                        <div bgcolor="#FFFFFF" text="#000000" class="">
                          <br class="">
                          <br class="">
                          Looks like problem started around 2016-04-17
                          20:19:34,822, based on engine.log attached.<br
                            class="">
                        </div>
                      </div>
                    </blockquote>
                    <div><br class="">
                    </div>
                    <div>yes, that time looks correct. Any idea what
                      might have been a trigger? Anything interesting
                      happened at that time (power outage of some host,
                      some maintenance action, anything)? </div>
                    <div>logs indicate a problem when vdsm talks to
                      libvirt(all those "monitor become unresponsive”)</div>
                    <div><br class="">
                    </div>
                    <div>It does seem that at that time you started to
                      have some storage connectivity issues - first one
                      at 2016-04-17 20:06:53,929. And it doesn’t look
                      temporary because such errors are still there
                      couple hours later(in your most recent file you
                      attached I can see at 23:00:54)</div>
                    <div>When I/O gets blocked the VMs may experience
                      issues (then VM gets Paused), or their qemu
                      process gets stuck(resulting in libvirt either
                      reporting error or getting stuck as well -&gt;
                      resulting in what vdsm sees as â€œmonitor
                      unresponsive”)</div>
                    <div><br class="">
                    </div>
                    <div>Since you now bounced libvirtd - did it help?
                      Do you still see wrong status for those VMs and
                      still those "monitor unresponsive" errors in
                      vdsm.log?</div>
                    <div>If not…then I would suspect the â€œvm
                      recovery” code not working correctly. Milan is
                      looking at that.</div>
                    <div><br class="">
                    </div>
                    <div>Thanks,</div>
                    <div>michal</div>
                    <div>
                      <div><br class="">
                      </div>
                    </div>
                    <div class=""><br class="">
                    </div>
                    <blockquote type="cite" class="">
                      <div class="">
                        <div bgcolor="#FFFFFF" text="#000000" class="">
                          There's a lot of vdsm logs!<br class="">
                          <br class="">
                          fyi, the storage domain for these Vms is a
                          "local" nfs share,
                          7e566f55-e060-47b7-bfa4-ac3c48d70dda.<br
                            class="">
                          <br class="">
                          attached more logs.<br class="">
                          <br class="">
                          <br class="">
                          <div class="moz-cite-prefix">On 04/28/2016
                            12:53 AM, Michal Skrivanek wrote:<br
                              class="">
                          </div>
                          <blockquote
                            cite="mid:28BF55E6-3A90-4BB7-90B9-1EE0A82FC460@redhat.com"
                            type="cite" class="">
                            <blockquote type="cite" class="">
                              <pre class="" wrap="">On 27 Apr 2016, at 19:16, Bill James <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:bill.james@j2.com">&lt;bill.james@j2.com&gt;</a> wrote:

virsh # list --all
error: failed to connect to the hypervisor
error: no valid connection
error: Failed to connect socket to '/var/run/libvirt/libvirt-sock': No such file or directory

</pre>
                            </blockquote>
                            <pre class="" wrap="">you need to run virsh in read-only mode
virsh -r list â€”all

</pre>
                            <blockquote type="cite" class="">
                              <pre class="" wrap="">[root@ovirt1 test vdsm]# systemctl status libvirtd
● libvirtd.service - Virtualization daemon
  Loaded: loaded (/usr/lib/systemd/system/libvirtd.service; enabled; vendor preset: enabled)
 Drop-In: /etc/systemd/system/libvirtd.service.d
          â””─unlimited-core.conf
  Active: active (running) since Thu 2016-04-21 16:00:03 PDT; 5 days ago


tried systemctl restart libvirtd.
No change.

Attached vdsm.log and supervdsm.log.


[root@ovirt1 test vdsm]# systemctl status vdsmd
● vdsmd.service - Virtual Desktop Server Manager
  Loaded: loaded (/usr/lib/systemd/system/vdsmd.service; enabled; vendor preset: enabled)
  Active: active (running) since Wed 2016-04-27 10:09:14 PDT; 3min 46s ago


vdsm-4.17.18-0.el7.centos.noarch
</pre>
                            </blockquote>
                            <pre class="" wrap="">the vdsm.log attach is good, but it’s too short interval, it only shows recovery(vdsm restart) phase when the VMs are identified as paused….can you add earlier logs? Did you restart vdsm yourself or did it crash?


</pre>
                            <blockquote type="cite" class="">
                              <pre class="" wrap="">libvirt-daemon-1.2.17-13.el7_2.4.x86_64


Thanks.


On 04/26/2016 11:35 PM, Michal Skrivanek wrote:
</pre>
                              <blockquote type="cite" class="">
                                <blockquote type="cite" class="">
                                  <pre class="" wrap="">On 27 Apr 2016, at 02:04, Nir Soffer <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:nsoffer@redhat.com">&lt;nsoffer@redhat.com&gt;</a> wrote:

jjOn Wed, Apr 27, 2016 at 2:03 AM, Bill James <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:bill.james@j2.com">&lt;bill.james@j2.com&gt;</a> wrote:
</pre>
                                  <blockquote type="cite" class="">
                                    <pre class="" wrap="">I have a hardware node that has 26 VMs.
9 are listed as "running", 17 are listed as "paused".

In truth all VMs are up and running fine.

I tried telling the db they are up:

engine=&gt; update vm_dynamic set status = 1 where vm_guid =(select
vm_guid from vm_static where vm_name = '<a moz-do-not-send="true" href="http://api1.test.j2noc.com" class="">api1.test.j2noc.com</a>');

GUI then shows it up for a short while,

then puts it back in paused state.

2016-04-26 15:16:46,095 INFO [org.ovirt.engine.core.vdsbroker.VmAnalyzer]
(DefaultQuartzScheduler_Worker-16) [157cc21e] VM '242ca0af-4ab2-4dd6-b515-5
d435e6452c4'(<a moz-do-not-send="true" href="http://api1.test.j2noc.com" class="">api1.test.j2noc.com</a>) moved from 'Up' --&gt; 'Paused'
2016-04-26 15:16:46,221 INFO [org.ovirt.engine.core.dal.dbbroker.auditlogh
andling.AuditLogDirector] (DefaultQuartzScheduler_Worker-16) [157cc21e] Cor
relation ID: null, Call Stack: null, Custom Event ID: -1, Message: VM api1.
<a moz-do-not-send="true" href="http://test.j2noc.com" class="">test.j2noc.com</a> has been paused.


Why does the engine think the VMs are paused?
Attached engine.log.

I can fix the problem by powering off the VM then starting it back up.
But the VM is working fine! How do I get ovirt to realize that?
</pre>
                                  </blockquote>
                                  <pre class="" wrap="">If this is an issue in engine, restarting engine may fix this.
but having this problem only with one node, I don't think this is the issue.

If this is an issue in vdsm, restarting vdsm may fix this.

If this does not help, maybe this is libvirt issue? did you try to check vm
status using virsh?
</pre>
                                </blockquote>
                                <pre class="" wrap="">this looks more likely as it seems such status is being reported
logs would help, vdsm.log at the very least.

</pre>
                                <blockquote type="cite" class="">
                                  <pre class="" wrap="">If virsh thinks that the vms are paused, you can try to restart libvirtd.

Please file a bug about this in any case with engine and vdsm logs.

Adding Michal in case he has better idea how to proceed.

Nir
</pre>
                                </blockquote>
                              </blockquote>
                              <pre class="" wrap=""><a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:Users@ovirt.org">Users@ovirt.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/listinfo/users</a>
</pre>
                            </blockquote>
                          </blockquote>
                          <br class="">
                        </div>
                        <span
                          id="cid:EB02B488-C070-46FA-9938-DC7D6DF5BEED@brq.redhat.com">&lt;engine.log-20160421.gz&gt;</span><span
id="cid:52E27023-A602-4DB0-B69A-18237CC048A3@brq.redhat.com">&lt;vdsm.logs.tar.gz&gt;</span></div>
                    </blockquote>
                  </div>
                  <br class="">
                </blockquote>
                <br>
                <p><a moz-do-not-send="true"
href="http://www.j2.com/?utm_source=j2global&amp;utm_medium=xsell-referral&amp;utm_campaign=employeeemail"><span
                      style="color:windowtext;
                      text-decoration:none"><img moz-do-not-send="true"
src="http://home.j2.com/j2_Global_Cloud_Services/j2_Global_Email_Footer.jpg"
                        alt="www.j2.com" height="46" border="0"
                        width="391"></span></a></p>
                <p><span
style="font-size:8.0pt;font-family:&quot;Arial&quot;,&quot;sans-serif&quot;;color:gray">This
                    email, its contents and attachments contain
                    information from <a moz-do-not-send="true"
href="http://www.j2.com/?utm_source=j2global&amp;utm_medium=xsell-referral&amp;utm_campaign=employemail">j2
                      Global, Inc</a>. and/or its affiliates which may
                    be privileged, confidential or otherwise protected
                    from disclosure. The information is intended to be
                    for the addressee(s) only. If you are not an
                    addressee, any disclosure, copy, distribution, or
                    use of the contents of this message is prohibited.
                    If you have received this email in error please
                    notify the sender by reply e-mail and delete the
                    original message and any copies. © 2015 <a
                      moz-do-not-send="true" href="http://www.j2.com/">j2
                      Global, Inc</a>. All rights reserved. <a
                      moz-do-not-send="true" href="http://www.efax.com/">eFax
                      ®</a>, <a moz-do-not-send="true"
                      href="http://www.evoice.com/">eVoice ®</a>, <a
                      moz-do-not-send="true"
                      href="http://www.campaigner.com/">Campaigner ®</a>,
                    <a moz-do-not-send="true"
                      href="http://www.fusemail.com/">FuseMail ®</a>, <a
                      moz-do-not-send="true"
                      href="http://www.keepitsafe.com/">KeepItSafe ®</a>
                    and <a moz-do-not-send="true"
                      href="http://www.onebox.com/">Onebox ®</a> are !
                    registere d trademarks of <a moz-do-not-send="true"
                      href="http://www.j2.com/">j2 Global, Inc</a>. and
                    its affiliates.</span></p>
              </div>
            </blockquote>
          </div>
        </div>
      </div>
    </blockquote>
    <br>
  
<p><a href="http://www.j2.com/?utm_source=j2global&utm_medium=xsell-referral&utm_campaign=employeeemail"><span style='color:windowtext;
text-decoration:none'><img border=0 width=391 height=46
src="http://home.j2.com/j2_Global_Cloud_Services/j2_Global_Email_Footer.jpg" alt="www.j2.com"></span></a></p>

<p><span style='font-size:8.0pt;font-family:"Arial","sans-serif";
color:gray'>This email, its contents and attachments contain information from <a href="http://www.j2.com/?utm_source=j2global&utm_medium=xsell-referral&utm_campaign=employemail">j2 Global, Inc</a>. and/or its affiliates which may be privileged, confidential or otherwise protected from disclosure. The information is intended to be for the addressee(s) only. If you are not an addressee, any disclosure, copy, distribution, or use of the contents of this message is prohibited. If you have received this email in error please notify the sender by reply e-mail and delete the original message and any copies.  2015 <a href="http://www.j2.com/">j2 Global, Inc</a>. All rights reserved. <a href="http://www.efax.com/">eFax </a>, <a href="http://www.evoice.com/">eVoice </a>, <a href="http://www.campaigner.com/">Campaigner </a>, <a href="http://www.fusemail.com/">FuseMail </a>, <a href="http://www.keepitsafe.com/">KeepItSafe </a> and <a href="http://www.onebox.com/">Onebox </a> are registered trademarks of <a href="http://www.j2.com/">j2 Global, Inc</a>. and its affiliates.</span></p></body>
</html>