On Wed, Mar 29, 2017 at 4:35 PM, Nicolas Ecarnot <nicolas@ecarnot.net> wrote:

[Please ignore the previous msg]

Hello,

One of our DC is a very small one, though quite critical.
It's almost hyper converged : hosts are compute+storage, but the engine is standalone.

It's made of :

Hardware :
- one physical engine : CentOS 6.7
- 3 physical hosts : CentOS 7.2

Software :
- oVirt 3.6.5
- glusterFS 3.7.16 in replica-3, sharded.

The goal is to upgrade all this to oVirt 4.1.1, and also upgrade the OSes. (oV 4.x only available on cOS 7.x)

At present, only 3 VMs here are critical, and I have backups for them.
Though, I'm quite nervous with the path I have to follow and the hazards. Especially about the gluster parts.

At first glance, I would go this way (feel free to comment) :
- upgrade the OS of the engine : 6.7 -> 7.3
- upgrade the OS of the hosts  : 7.2 -> 7.3
- upgrade and manage the upgrade of gluster, check the volumes...
- upgrade oVirt (engine then hosts)

Hi Nicolas,

Regards the gluster update - have your tried to ask in the gluster ml?

I'm adding Sahina to share her opinion on that flow. 

Export domain as Didi suggested is indeed the easy way from oVirt perspective - but it'll take a long time to copy/restore, Gluster backup should perform faster if there's a solution that fit your needs.

But when upgrading the OSes, I guess it will also upgrade the gluster layer.

During this upgrade, I have no constraint to keep everything running, total shutdown is acceptable.

Is the above procedure seems OK, or may am I missing some essential points?

Thank you.

--
Nicolas ECARNOT
_______________________________________________
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