Node/IP: <name> is not a known host of the cluster.

[bubaphex]

Hi all,

I am trying to separate node from a cluster(without reinstalling) and I am following these instructions: link here

When I get to

pvecm delnode oldnode

I get this error

Node/IP: beast is not a known host of the cluster.

beast being the name of the old node

I've googled around but can't find anything to solve the issue

any help is much appreciated

 

[fiona]

Hi,
does the node appear in your /etc/pve/corosync.conf? What is the output of pvecm status on the node to be removed and on some other cluster node?

 

[oscar.martinez]

have the same problem, the node is still in /etc/pve/corosync.conf but in with the command pvecm status, i got just the two nodes are good, I wanted to change the ip of one node like the indication here : https://forum.proxmox.com/threads/change-cluster-nodes-ip-addresses.33406/post-163795

 

[fiona]

Hi,
did you increment the config_version when you edited the corosync configuration: https://pve.proxmox.com/pve-docs/chapter-pvecm.html#_corosync_configuration ? If the corosync config does not match between nodes, see: https://forum.proxmox.com/threads/corosync-cluster-problems-help.55856/post-257275

 

[oscar.martinez]

Hi,
I didn't increment the config version and also now the corosync are different. Currently, two of the nodes are synchronized, but the third node (node3) is missing from the 'corosync.conf' file on these nodes. Additionally, in the host file, only localhost and host IP is listed, not the IPs from the other members of the cluster.
In the web browser interface's tree menu for node1 and node2, node3 is still visible (marked as not reachable), even though it doesn't exist in the cluster menu.
in the node3, , I have the other two nodes listed in both the 'corosync.conf' and 'hosts' files. In the web browser are also visible (but not reachable) and also presents in the menu cluster.
Could be possible to take out the node3 from the cluster (node1 and node2 working) and then reinstall node3 and again add it to the cluster?

 

[fiona]

Hi,
I didn't increment the config version and also now the corosync are different. Currently, two of the nodes are synchronized, but the third node (node3) is missing from the 'corosync.conf' file on these nodes. Additionally, in the host file, only localhost and host IP is listed, not the IPs from the other members of the cluster.

Please post the output of pvecm status. Did you maybe already run pvecm delnode <NODE> for the old node? In that case it might be that just the configuration directory is still there. Then you can just re-install and re-join.

But you still might want to increment the config version, just to be sure. For how to edit the configuration safely, see: https://pve.proxmox.com/pve-docs/chapter-pvecm.html#_corosync_configuration

Could be possible to take out the node3 from the cluster (node1 and node2 working) and then reinstall node3 and again add it to the cluster?

Yes: https://pve.proxmox.com/pve-docs/chapter-pvecm.html#_remove_a_cluster_node
Alternatively, if you didn't run pvecm delnode <NODE> yet, you could also try and follow: https://forum.proxmox.com/threads/corosync-cluster-problems-help.55856/post-257275

 

[oscar.martinez]

pvecm status on one of the nodes are working:
Cluster information
-------------------
Name: pve-cluster
Config Version: 10
Transport: knet
Secure auth: on

Quorum information
------------------
Date: Thu Mar 7 11:51:13 2024
Quorum provider: corosync_votequorum
Nodes: 2
Node ID: 0x00000002
Ring ID: 1.abc
Quorate: Yes

Votequorum information
----------------------
Expected votes: 2
Highest expected: 2
Total votes: 2
Quorum: 2
Flags: Quorate

Membership information
----------------------
Nodeid Votes Name
0x00000001 1 176.XXX.XXX.XXX
0x00000002 1 136.XXX.XXX.XXX (local)

pvecm status on the node is not working:
Cluster information
-------------------
Name: pve-cluster
Config Version: 9
Transport: knet
Secure auth: on

Quorum information
------------------
Date: Thu Mar 7 11:51:11 2024
Quorum provider: corosync_votequorum
Nodes: 1
Node ID: 0x00000003
Ring ID: 3.ad1
Quorate: No

Votequorum information
----------------------
Expected votes: 3
Highest expected: 3
Total votes: 1
Quorum: 2 Activity blocked
Flags:

Membership information
----------------------
Nodeid Votes Name
0x00000003 1 88.XXX.XXX.XXX (local)

I already run the command pvecm delnode <NODE> and I got the message:

Node/IP: 3 is not a known host of the cluster.

if I try to create or modify the corosync file on the server with problems :
cp /etc/pve/corosync.conf /etc/pve/corosync.conf.bak
cp: cannot create regular file '/etc/pve/corosync.conf.bak': Permission denied

 

[fiona]

pvecm status on one of the nodes are working:
Cluster information
-------------------
Name: pve-cluster
Config Version: 10
Transport: knet
Secure auth: on

Quorum information
------------------
Date: Thu Mar 7 11:51:13 2024
Quorum provider: corosync_votequorum
Nodes: 2
Node ID: 0x00000002
Ring ID: 1.abc
Quorate: Yes

Votequorum information
----------------------
Expected votes: 2
Highest expected: 2
Total votes: 2
Quorum: 2
Flags: Quorate

Membership information
----------------------
Nodeid Votes Name
0x00000001 1 176.XXX.XXX.XXX
0x00000002 1 136.XXX.XXX.XXX (local)

Looks like the the node already got deleted (Highest expected: 2). You can re-install the third node and re-join.

 

[oscar.martinez]

And what should I do to make not visible the server that already is out, on the web site in the node tree explorer at left ?

 

[fiona]

Please read the documentation: https://pve.proxmox.com/pve-docs/chapter-pvecm.html#_remove_a_cluster_node

The configuration files for the removed node will still reside in/etc/pve/nodes/hp4. Recover any configuration you still need and remove thedirectory afterwards.

 

[oscar.martinez]

Ok ready, it works, thanks a lot