completely removing a failed cluster-node

[hk@]

Hi,
in a pve7.4 cluster setup (4 nodes) one sadly died.
Before final "delnode" removed all entries from /etc/pve/replication.cfg that had this node as a target and re-grouped the replication to the remaining nodes.
Then regrouped all replication-groups, only then according to https://pve.proxmox.com/pve-docs/pve-admin-guide.html#_remove_a_cluster_node removed the node by "pvecm delnode".

pvecm is happy now, the node is gone, the webgui however still shows the node in the datacenter-tree and in the HA section it is still shown as a type LRM, however it is noted the timestamp is old and probably dead.

My curios question would be: how to remove the node from all these sections?

thx
hk

 

[CalebSnell]

I just had this same problem, a node called "pve-htpc" was stuck in the webgui and pvecm delnode pve-htpc said the node was not found in the cluster.

I checked /etc/pve/nodes/ and there was a folder for that node. I ran "rm -rf pve-htpc" from that folder and after a few seconds it was gone from the GUI.

Be very very careful entering this command, make sure it's written properly, double - triple check before pressing enter. Possible even backup the folder to another folder outside of /etc/ just in case you need to move it back.