[ovirt-users] Replace one engine host
Davide Ferrari
davide at billymob.com
Sun Sep 18 17:21:44 EDT 2016
Oh, thanks! You put me on the right way! Now it works:
Steps that I did:
- reinstall + deploy vm03
- manually set from maintenance to none on vm03
- reinstall + undeploy vm04 (I left it deployed again last time, trying to
remove it)
Then I found the problem: vm03 hosted-engine conf
(/etc/ovirt-hosted-engine/hosted-engine.conf) had host_id set to 4 which
conflicted with vm04 host_id. So I changed it to 2 (the only one free),
restarted ovirt-ha-agent (and broker just to be sure) and it worked! Now
from vm-status I see vm01,02 and 03 with 3400 as priority and I can migrate
the HE vm to any of them!
Thanks a lot for your kind help, Roy!
2016-09-18 22:00 GMT+02:00 Roy Golan <rgolan at redhat.com>:
> Can you share the deploy logs of vm03? /var/log/ovirt-engine/host-deploy/
> Is there an /etc/ovirt-hosted-engine/hosted-engine.conf at all?
>
> On 18 September 2016 at 17:28, Davide Ferrari <davide at billymob.com> wrote:
>
>> Hello Roy, I forgot to mention that I already reinstalled (several times)
>> vm03 from the gui with the deploy option checked, but there's no way it
>> appears there. For the record vm04 was the first host deployed (from cli)
>> after the initial vm01 installation, then vm02 and then 03.
>>
>> On Sep 18, 2016 11:06, "Roy Golan" <rgolan at redhat.com> wrote:
>>
>>> vm03 is missing here from the status. Can you reinstall it from UI (if
>>> you added the host vm03 then just go to Management menu and click
>>> reinstall, choose 'Deploy' Option) - After its done, when the host is
>>> activated you'll need to manually switch the host out of maintenence (we
>>> have a bug on it - solved in master already)
>>>
>>>
>>>
>>> On 18 September 2016 at 10:45, Davide Ferrari <davide at billymob.com>
>>> wrote:
>>>
>>>> Hello
>>>>
>>>> Still playing a bit with oVirt, my cluster is ovirt+gluster with a
>>>> replica 3 glusterfs for the engine (on vm01, vm02 and vm03) and 4 hosts for
>>>> the data glusterfs and running virtual machines (vm01, vm02, vm03 and
>>>> vm04). Now, by mistake, I deployed hosted-engine on vm04 but I don't want
>>>> to have the engine on a machine without local data + it's not working
>>>> correctly (weight 0 and marked as in maintenance). Now, I've already tried
>>>> to reinstall it from the the GUI marking the "Undeploy" check, reinstall
>>>> vm03 marking the "deploy" check but nothing seems to work. Here it is the
>>>> hosted-engine --vm-status output from vm04. What can I do to deploy
>>>> correctly the hosted engine on vm03?
>>>>
>>>> --== Host 1 status ==--
>>>>
>>>> Status up-to-date : True
>>>> Hostname : vm01.mydomain.tld
>>>> Host ID : 1
>>>> Engine status : {"reason": "vm not running on this
>>>> host", "health": "bad", "vm": "down", "detail": "unknown"}
>>>> Score : 3400
>>>> stopped : False
>>>> Local maintenance : False
>>>> crc32 : fee2f7d8
>>>> Host timestamp : 48911
>>>> Extra metadata (valid at timestamp):
>>>> metadata_parse_version=1
>>>> metadata_feature_version=1
>>>> timestamp=48911 (Sat Sep 17 07:57:02 2016)
>>>> host-id=1
>>>> score=3400
>>>> maintenance=False
>>>> state=EngineDown
>>>> stopped=False
>>>>
>>>>
>>>> --== Host 3 status ==--
>>>>
>>>> Status up-to-date : True
>>>> Hostname : vm02.mydomain.tld
>>>> Host ID : 3
>>>> Engine status : {"health": "good", "vm": "up",
>>>> "detail": "up"}
>>>> Score : 3400
>>>> stopped : False
>>>> Local maintenance : False
>>>> crc32 : 9138d24e
>>>> Host timestamp : 48907
>>>> Extra metadata (valid at timestamp):
>>>> metadata_parse_version=1
>>>> metadata_feature_version=1
>>>> timestamp=48907 (Sat Sep 17 07:57:00 2016)
>>>> host-id=3
>>>> score=3400
>>>> maintenance=False
>>>> state=EngineUp
>>>> stopped=False
>>>>
>>>>
>>>> --== Host 4 status ==--
>>>>
>>>> Status up-to-date : False
>>>> Hostname : vm04.mydomain.tld
>>>> Host ID : 4
>>>> Engine status : unknown stale-data
>>>> Score : 0
>>>> stopped : False
>>>> Local maintenance : True
>>>> crc32 : 221d262e
>>>> Host timestamp : 17958
>>>> Extra metadata (valid at timestamp):
>>>> metadata_parse_version=1
>>>> metadata_feature_version=1
>>>> timestamp=17958 (Fri Sep 16 23:19:32 2016)
>>>> host-id=4
>>>> score=0
>>>> maintenance=True
>>>> state=LocalMaintenance
>>>> stopped=False
>>>>
>>>>
>>>> --
>>>> Davide Ferrari
>>>> Senior Systems Engineer
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at ovirt.org
>>>> http://lists.ovirt.org/mailman/listinfo/users
>>>>
>>>>
>>>
>
--
Davide Ferrari
Senior Systems Engineer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/users/attachments/20160918/7435f8d5/attachment-0001.html>
More information about the Users
mailing list