On Fri, Aug 19, 2016 at 8:11 PM, Scott <romracer@gmail.com> wrote:

Hi Sandvik,

I believe this ultimately is what the self hosted engine is planned to solve. It's up to you as to whether you feel comfortable with that idea currently.

Indeed. And many in the community are using it.

Outside of that, I would point you to the Red Hat Cluster Suite in a RHEV/RHEL environment. The high availability component of that is based on Linux-HA but I have only used Red Hat's implementation.

Lastly, the poor man's solution if this isn't critical infrastructure would be a hot standby server. Keep the packages in sync between the two however you feel appropriate. Run engine-backup as frequently as needed and copy it to the second server. Have your engine address as a secondary address with appropriate DNS and move it between them in a fail over (or use a CNAME). Restore the backup in a fail over and run engine-setup. I think you get the idea.

I don't like this approach very much - you always forget some configuration. (For example, we've had an issue with Kerberos database later on when migrating).

You are also 'wasting' a server.

Personally I use hosted engine in all but my main production cluster. That one will use it once I've upgraded those hosts to RHEL 7.

Hope that helps.

Scott

On Fri, Aug 19, 2016, 11:40 AM Sandvik Agustin <agustinsandvik@gmail.com> wrote:
Hi Ovirt Users,

Do you have any documentation, tutorial, howto or workaround to perform or to have an ovirt-engine HA? Iv'e tried to google about this but I can't find any.

let's say I want to have two engine, engine1 is currently managing two hypervisor, then engin2 is my reserved engine, so I can still managed those vm that are running inside of my hypervisor if ever engine1 fails.

TIA :)
Sandvik
_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users