Thanks for the report! :-)
On Wed, Jan 2, 2019 at 4:47 PM maoz zadok <maozza(a)gmail.com> wrote:
Host restart resolve the issue!
On Wed, Jan 2, 2019 at 11:50 AM maoz zadok <maozza(a)gmail.com> wrote:
>
> Thank you Didi for the fast reply!
> The problem was before the restore from backup (that was the reason that I reboot the
engine and the engine didn't come up, I dug a bigger hole :-( ),
> HA is not marked. but The VM was trying to migrate from kvm7 to kvm5 before the
engine crashed.
> I don't see the qemu process on any of the hosts.
> "vdsm-client Host getVMList fullStatus=True" show the machine uuid only on
one (kvm5)
>
> I'll try restarting the host (kvm5), let's see if it helps.
>
>
>
>
>
> On Wed, Jan 2, 2019 at 11:07 AM Yedidyah Bar David <didi(a)redhat.com> wrote:
>>
>> On Wed, Jan 2, 2019 at 10:51 AM maoz zadok <maozza(a)gmail.com> wrote:
>> >
>> > Hello,
>> > vm stuck (on two hosts)
>> > I have MySQL VM guest that was stuck in a way that "kill -9" to
the MySQL server pid did nothing.
>> > from the shell I execute "init 0", this did not shut down the VM.
>> >
>> > from the ovirt-engine I'm trying to power off the machine, and it is not
doing anything.
>> > the event shows:
>> >>
>> >> VDSM kvm5 command DestroyVDS failed: Virtual machine destroy error
>> >
>> > and:
>> >>
>> >> Failed to power off VM mysql-production (Host: kvm5, User:
admin@internal-authz).
>> >
>> >
>> > Note, when I list the VM s on the hosts view, I see that that same VM is on
two hosts (kvm7 and kvm5).
>> > It looks like the VM is stuck on two hosts, in the status "Powering
Down"
>> >
>> > any ideas?
>>
>> No idea about this specific problem and how to solve it.
>>
>> Is this on the same system you restored from backup? Is this VM marked
>> as HA (highly-available)? Did it move to a different host since you
>> took the backup you now restored?
>>
>> If all are 'yes', then it's likely that the engine tried to start it
>> (because it's HA) before it noticed that it's already running on
>> another host (a fact that it could not know, if it was different from
>> what's in the restored database). Sadly, this very likely leads to
>> corruption. See also:
>>
>>
https://bugzilla.redhat.com/1441322
>>
>> Since I do not see 'Resetting HA VM status' in the restore logs you
posted,
>> it might be a different issue, but perhaps similar.
>>
>> If indeed you see two qemu processes on the hosts, I'd try to kill -9
>> both of them.
>>
>> Adding Arik.
>>
>> Good luck,
>>
>>
>> > Thank you
>> > _______________________________________________
>> > Users mailing list -- users(a)ovirt.org
>> > To unsubscribe send an email to users-leave(a)ovirt.org
>> > Privacy Statement:
https://www.ovirt.org/site/privacy-policy/
>> > oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
>> > List Archives:
https://lists.ovirt.org/archives/list/users@ovirt.org/message/E2HKUJZYUNC...
>>
>>
>>
>> --
>> Didi