Closed croneter closed 2 months ago
You specified all nodes in talosctl config nodes
. It is not recommended. So you get a mix of results from multiple nodes at once
Ok thanks @smira. How do I fix that?
https://www.talos.dev/v1.7/introduction/getting-started/#understand-talosctl-endpoints-and-nodes
The easiest you can do talosctl config node A.B.C.D
Sorry @smira, this is not really helping yet. I'm trying to re-join 192.168.50.52 (see above, removed the node from the cluster). But talosctl apply-config -n 192.168.50.52 --file controlplane.yaml --insecure
is not working: after a reboot, the node returns to a Stage: Maintenance
Could you be so kind and elaborate a bit more?
Let's not mix several things as one issue.
Please open a discussion with a question, but if apply-config
doesn't work and the machine returns to maintenance, there should be something in the logs which explains why, but I can't guess.
I'm going to close this issue, please open a new one if the problem is still there.
Bug Report
I somehow butched something with Talos, don't ask me how :-(. Nodes now show up 3x in both
talosctl get members
andtalosctl etcd members list
. The Kubernetes cluster seems to work fine in principle. BUT I e.g. just upgraded Talos from 1.7.5 to 1.7.6, that was a pain as the update failed repeatetly, probably because the wrong node entry was selected among the 3 entries at random.Description
Let me know what else you might need.
Node appearing three times in Talos and etcd:
Removing a node with
talosctl -n 192.168.50.52 reset
results in errors that 192.168.50.52 not being found - without the ability to e.g. delete it in etcd.Logs
etcd log attached: etcd.log.txt
Environment