OVIRT Engine Restore from old Backup
Hey All, I really dont like spamming the forums but the issues I am having keep changing entirely. I am at the point where I can get much closer to rebuilding my engine although my hurtle currently is restoring. When I restore using my most recent backup it is problematic Currently I am restoring by: clean engine install engine-cleanup engine-backup --mode=restore --file=ovirt-engine-backup-20220504011012.backup --log=restore5.log engine-setup Comes up with alot of annoyances as the ID of the old hosted_storage and the new one are mismatched in the GUI. On top of that it is not allowing me to deploy the engine to the 2nd Gluster (3rd one keeps deleting its glusters on reboot since 4.3 Upgrade). Host 1 which runs the hosted engine is not showing as active or unlike, rather it is unresponsive, when I tried to enroll certificates it finally accepted but failed because it said its missing networks, once added it crashed the engine and ha agent wouldnt come up. Now that I think of it, I could have removed the extra networks from being required although it still doesnt benefit me here as I am unable to deploy or reinstall the 2nd Gluster node and the first one will not show its correct state as the host. Should I just be deploying the DB, or is the full file backup fine? My issue here is it takes long to redeploy so the trial an error of failing takes me hours as my backup is from 4.5 engine even though my nodes are on 4.4 as 4.5 keeps causing the catastrophic failure of my whole cluster.
I have even tried usine answers files when deploying the engine with the backup and marking what I want the UUID to be and the name of the domain but it continues to fail.
In worst case scenario, you can import the Storage domains ,the VMs &Templates from it and then configure host networking. Keep in mind that 2 Engines must never manage the same host. Best Regards,Strahil Nikolov On Sat, May 7, 2022 at 10:15, Abe E<aellahib@gmail.com> wrote: I have even tried usine answers files when deploying the engine with the backup and marking what I want the UUID to be and the name of the domain but it continues to fail. _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/2YWWVLPTYKCXCN...
So I used my most recent backup I could make work (March but better than nothing). Got all the glusters going and the hosts, my only issue currently is the data domain was imported with the restore deployment despite this it shows offline and inactive. Storage ID in GUI shows as e9c41992-2130-4ce8-96fa-39be109c2b96, likewise same DIR name in gluster_bricks in CLI, but it shows N/A for size. It also says failed : Storage Domain data (Data Center LabNetDev) was deactivated by system because it's not visible by any of the hosts. Failed to activate Storage Domain data (Data Center In the actual mount though, the directory is empty. So if i cat /rhev/data-center/mnt/glusterSD/gluster-1.removed.com:_data it is empty. Do you know why this would be?
Spoke too soon, the above + it seems occasionally the gluster on 2 nodes reports Cluster quorum is not met. I did have to at one point rebuild the glusters which includes the bricks, am I at the point where I should be resetting the brick on a node?
If you decide to do it, I would do the following: Verify if there are any heals towards the 4.4 node -> If not, proceed.use remove-brick replica 1 mode (see gluster volume help) and remove both 4.5 bricks.Then remove the nodes with gluster peer detach.Remove gluster 4.5 and install 4.4 Then setup only 1 node's brick (mkfs.xfs-f -i size=512 ...) , peer probe that node and expand the volume (add-brick replica 2) . N.B.: Don't forget to increase the replica count !!! Wait for the heals to end, peer probe the last node, mkfs.xfs the brick and then again add-brick and increase the replica count. Of course, backups are must when you play with storage :D Best Regards,Strahil Nikolov On Tue, May 10, 2022 at 1:01, Abe E<aellahib@gmail.com> wrote: Spoke too soon, the above + it seems occasionally the gluster on 2 nodes reports Cluster quorum is not met. I did have to at one point rebuild the glusters which includes the bricks, am I at the point where I should be resetting the brick on a node? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/U6HJD4NH334HFZ...
Do you think I could salvage the data at this point? I have only node that remains saying it has 1.5TB of 4/5ishTB although I cannot find within the image folders my VMsize that I expect. I am beginning to think my gluster had issues before 4.5 "broke" it. 27K Unsynced Entries doesnt sound right, I would have expected it to be healing itself. I am close to giving up, even after rebuilding the bricks Also, thank you again for your repeated help
uczestnicy (2)
-
Abe E -
Strahil Nikolov