Replies: 1 comment 3 replies
-
|
Are these the steps you followed to restore the cluster backup? https://omni.siderolabs.com/how-to-guides/restore-etcd-of-a-cluster-managed-by-cluster-templates |
Beta Was this translation helpful? Give feedback.
3 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment

Uh oh!
There was an error while loading. Please reload this page.
-
I have been running a personal homelab cluster for over a month, with a single node (n1) control plane on Omni, with daily backups configured. I was dealing with some issues with Ceph and trying to do a Ceph Cluster restore, but due to the Ceph cluster issue, the k8s-api was getting hit pretty hard and so was dealing with intermittent outages.
I was then trying to do a swap out and change the machines in the cluster (a1, a2, a3) being CPs and n1 switching over to worker.
Typically, I would do omnictl delete machineset -control-planes but this cluster was stuck in deletion and I accidentally ended up deleting the cluster itself (big mistake).
So now when I tried to do omnictl cluster template sync -f - it isn’t able to validate the cluster UUID, since the cluster itself is deleted.
I then did rclone of the old cluster’s snapshot over to the new cluster’s object storage location so that it would exist, and modified the cluster template with the new UUID, but it wouldn’t take, likely due to the snapshot itself being recorded in omni’s etcd, since the error was: * error getting backup data: failed to get: resource BackupDatas.omni.sidero.dev(default/@undefined) doesn't exist
I know I should’ve backed up the etcd manually with talosctl and this would’ve made life so much easier, but now I’m stuck.
Would it be possible to restore the previous snapshot somehow?
Beta Was this translation helpful? Give feedback.
All reactions