Replies: 2 comments
-
Hello Sébastien, |
Beta Was this translation helpful? Give feedback.
0 replies
-
Hello Stéphan, Unfortunately we have found no other solution than to progressively move away from longhorn and use hosting service native volume management. cheers S. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi guys, here is the situation:
I have read the documentation about upgrading when using AWS/GCP/Azure managed services, but solutions there do not seem to apply to OVH case (for example i can not create a new node pool with new version of k8s, i'm stuck with current version).
My question here is: is there an (automatic) way to make longhorn use the newly seen disk, as if it was "added" ?
What I have tried with success is to manually change the disk UUID (in /var/lib/longhorn/longhorn-disk.cfg) using the old one, after each node upgrade. Then replicas are rebuilt as necessary, and the upgrade proceed correctly. Of course this is not usable in "real life", but it proves that an upgrade could be possible with no service disruption.
Thanks a lot for any idea about this,
cheers
Sebastien
Beta Was this translation helpful? Give feedback.
All reactions