https://rancher.com/ logo
Join Slack
Powered by
This message was deleted.
# general
a
This message was deleted.
p
You should restore from etcd snapshot
f
Restore Rancher, or the cluster we were attempting to upgrade?
p
the cluster sorry
f
I thought about that, but there are 54 deployments currently running on that cluster that are in use.
p
Ah yeah restoring from snapshot incurs downtime
f
The upgrade deployment has failed and no longer running on the cluster, but it still shows as upgrading in Rancher, and I still get the message "Loading" when I attempt to edit the cluster from the Rancher UI.
We resolved the issue today when we discovered some unusual characters when we viewed the yaml on the cluster. The value for spec.rke2Config.kubernetesVersion was v1.26.1s5+rke2r2. We couldn't edit the cluster, but I was able to edit the resource definition with
Copy code
kc edit <http://clusters.management.cattle.io|clusters.management.cattle.io> c-m-5gjnpjn4
where c-m-5gjnpjn4 is the cluster id where the problem existed. I was then replaced the value with the version number the cluster was running and save it. Upon saving the value, the "Upgraiding" state changed to Active, and I regained the capability to edit the config fo the cluster. We were then able to upgrade Kubernetes to 1.28 without further issues.
Open in Slack
PreviousNext
Powered by