Issues with cluster config

[Marcel Fuhrmann]

I've set up a new PVE instance, created the cluster configuration using the web UI, and installed a second PVE instance which should join the new created cluster.
Unfortunately the host entry of the second instance was wrong and pointed to a wrong IP address and after the join process the IP of the second server was wrong. Changing the IP of the server and in the corosync config didn't help, so I wanted to remove the cluster node again.

I've tried to remove the second cluster node using

```
pvecm expected 1
pvecm delnode 2
```

and

`pvecm nodes` only shows one nodes again.

But the web UI still has the second node listed with an error symbol on it.
I've reinstalled node 2 and wanted to join the node again but the web UI does not provide the cluster join information any more. I could create a new cluster using a new name, because PVE tells me the old name is still in use.

What's the correct approach to fix this issue?

Also there is a VM on node 1 running which should not be taken offline without knowing it

Thanks for your help!

 

[mira]

Is there anything left in /etc/pve/nodes/<node2>? This could explain why it still shows in the GUI.
Please post the output of 'pvecm status' and /etc/pve/corosync.conf

 

[Marcel Fuhrmann]

Is there anything left in /etc/pve/nodes/<node2>?

No

Please post the output of 'pvecm status' and /etc/pve/corosync.conf

You're right. The corosync.conf has still the node2 entry.

[12:32]root@fullipgvh-n2:~# pvecm status
Quorum information
------------------
Date:             Mon Feb 18 12:33:25 2019
Quorum provider:  corosync_votequorum
Nodes:            1
Node ID:          0x00000001
Ring ID:          1/32
Quorate:          Yes

Votequorum information
----------------------
Expected votes:   1
Highest expected: 1
Total votes:      1
Quorum:           1
Flags:            Quorate

Membership information
----------------------
    Nodeid      Votes Name
0x00000001          1 10.10.6.3 (local)
[12:33]root@fullipgvh-n2:~# cat /etc/pve/corosync.conf
logging {
  debug: off
  to_syslog: yes
}

nodelist {
  node {
    name: fullipgvh-n1
    nodeid: 2
    quorum_votes: 1
    ring0_addr: 10.10.6.2
  }
  node {
    name: fullipgvh-n2
    nodeid: 1
    quorum_votes: 1
    ring0_addr: 10.10.6.3
  }
}

quorum {
  provider: corosync_votequorum
}

totem {
  cluster_name: fullipgvh
  config_version: 2
  interface {
    bindnetaddr: 10.10.6.3
    ringnumber: 0
  }
  ip_version: ipv4
  secauth: on
  version: 2
}

PS: Don't be confused. the n2 labeled node is node 1 from PVE point of view and vice versa

 

[mira]

Looks like you'll have to manually remove it from the corosync config.

 

[Marcel Fuhrmann]

Looks like you'll have to manually remove it from the corosync config.

Just remove it and that's it?

 

[mira]

That should be it.

 

[Marcel Fuhrmann]

@dlimbeck Seems to work! Thanks!