Removing "main" node from cluster

[Petar Kozic]

Hi,
I checked the whole documentation about removing node from cluster and in general as I can see that is straightforward process.
But, I have one question becasue I can't find that in documentation.
I have 4 nodes in cluster, and I want to remove two, because that two which I want to remove is outdated and that is old servers.

pvesm status

Cluster information
-------------------
Name:             vcluster001
Config Version:   4
Transport:        knet
Secure auth:      on

Quorum information
------------------
Date:             Thu Sep 30 14:18:16 2021
Quorum provider:  corosync_votequorum
Nodes:            4
Node ID:          0x00000004
Ring ID:          1.24f9e
Quorate:          Yes

Votequorum information
----------------------
Expected votes:   4
Highest expected: 4
Total votes:      4
Quorum:           3
Flags:            Quorate

Membership information
----------------------
    Nodeid      Votes Name
0x00000001          1 public_ip
0x00000002          1 public_ip
0x00000003          1 public_ip
0x00000004          1 public_ip (local)

pvecm nodes

Membership information
----------------------
    Nodeid      Votes Name
         1          1 node004
         2          1 node005
         3          1 node008
         4          1 node009 (local)

So far, I logged in cluster over first node, like
https://node004:8006

Now, I want to remove that node004 and node005.

If I remove that nodes, how can I login in cluster when old "main" hostname will not be accessible anymore?
Does proxmox solve it itself when I remove nodes?

Thank you !

 

[oguz]

hi,

If I remove that nodes, how can I login in cluster when old "main" hostname will not be accessible anymore?
Does proxmox solve it itself when I remove nodes?

there's no "main" hostname, all PVE nodes are equal, you can access the other cluster nodes through the GUI of any cluster node

I have 4 nodes in cluster, and I want to remove two, because that two which I want to remove is outdated and that is old servers.

you can also do that one by one, remove node4 then upgrde, add back into cluster. do the same for node5

make sure you have backups of the containers/VMs and the necessary configuration files.

 

[Petar Kozic]

Thank you very much @oguz