[ovirt-users] Hosted Engine is down and won't start

[Simone Tiraboschi]

On Fri, Nov 10, 2017 at 9:42 AM, Kasturi Narra <knarra at redhat.com> wrote:

> Hello Logan,
>
>    One reason the liveliness check fails is host cannot ping your hosted
> engine vm. you can try connecting to HE vm using remote-viewer
> vnc://hypervisor-ip:5900 and from the hosted-engine --vm-status output
> looks like the HE vm is up and running fine.
>
>
Hi,
just a small addition:
we can deploy hosted-engine choosing vnc or spice as the graphical console
protocol so you have to fix the remote viewer command according to what you
are using.
And the tcp post is not always 5900 but it depends on the VMs starting
order.

To get the actual VNC port number you could use:
. /etc/ovirt-hosted-engine/hosted-engine.conf
vdsm-client VM getInfo vmID=$vmid | jq -r '.devices[] | select(.device |
contains("vnc")).port'

An alternative is to use the serial console with:
hosted-engine --console


>
>    - Please check internal dns setting like resolv.conf setting
>    - Can not resolve virtual host name or ip address.
>
> Thanks
> kasturi
>
>
> On Fri, Nov 10, 2017 at 12:56 PM, Logan Kuhn <support at jac-properties.com>
> wrote:
>
>> We lost the backend storage that hosts our self hosted engine tonight.
>> We've recovered it and there was no data corruption on the volume
>> containing the HE disk.  However, when we try to start the HE it doesn't
>> give an error, but it also doesn't start.
>>
>> The VM isn't pingable and the liveliness check always fails.
>>
>>  [root at ovirttest1 ~]# hosted-engine --vm-status | grep -A20 ovirttest1
>> Hostname                           : ovirttest1.wolfram.com
>> Host ID                            : 1
>> Engine status                      : {"reason": "failed liveliness
>> check", "health": "bad", "vm": "up", "detail": "up"}
>> Score                              : 3400
>> stopped                            : False
>> Local maintenance                  : False
>> crc32                              : 2c2f3ec9
>> local_conf_timestamp               : 18980042
>> Host timestamp                     : 18980039
>> Extra metadata (valid at timestamp):
>>        metadata_parse_version=1
>>        metadata_feature_version=1
>>        timestamp=18980039 (Fri Nov 10 01:17:59 2017)
>>        host-id=1
>>        score=3400
>>        vm_conf_refresh_time=18980042 (Fri Nov 10 01:18:03 2017)
>>        conf_on_shared_storage=True
>>        maintenance=False
>>        state=GlobalMaintenance
>>        stopped=False
>>
>> The environment is in Global Maintenance so that we can isolate it to
>> starting on a specific host to eliminate as many variables as possible.
>> I've attached the agent and broker logs
>>
>> _______________________________________________
>> Users mailing list
>> Users at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>>
>
> _______________________________________________
> Users mailing list
> Users at ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/users/attachments/20171113/c546b41a/attachment.html>