[SOLVED] Accidentally deleted corosync configs

[ninjabes]

I have accidentally deleted all corosync configs/files on one node of my cluster. Is there an easy way to get that node back into the cluster? I have already tried to copy /etc/corosync/* from another node but without success.

Thanks in advance!

 

[Moayad]

Hi,

Does the `/etc/pve/corosync.conf` is still exist on the deleted corosync node?

May you also tell us what exactly you deleted?

 

[ninjabes]

Hi,

Does the `/etc/pve/corosync.conf` is still exist on the deleted corosync node?

May you also tell us what exactly you deleted?

No, its no there anymore. I can only find a `/etc/pve/corosync.conf.bak`. The .bak file has all nodes included that should be in the cluster.

I wanted to delete a node from the cluster but I ssh'd into the wrong one. I used the following commands:

472   systemctl stop pve-cluster 
473   systemctl stop corosync 
474   pmxcfs -l 
475   rm /etc/pve/corosync.conf 
476   rm /etc/corosync/* 
477   rm -rf /etc/corosync/* 
478   killall pmxcfs 
479   systemctl start pve-cluster

 

[Moayad]

Hi,

I wanted to delete a node from the cluster but I ssh'd into the wrong one. I used the following commands:

Oops, :| that is a big mistake. In such a situation we recommend completely removing the node and back re-add it to the cluster again [0], hopefully, you have valid backups for the VMs and the important configs.


[0] https://pve.proxmox.com/pve-docs/pve-admin-guide.html

 

[ninjabes]

Ok I see! Yes, I have everything saved. Thanks for your help!

 

[ninjabes]

What if the node I have deleted was the first node of the cluster? Does this have any implications?

I wanted to remove the node from the cluster but currently its not working properly. The remaining cluster is complaining about quorum.

cluster not ready - no quorum?