Changed IPs on my cluster before reading anythings

M

mason64

Active Member
Apr 18, 2022
58
6
28
Hi,

I made a bit of a mistake :)

I changed my ips on my two nodes which were on a 10.0.0.x, for reasons i had to change my network to 192.168.1.x, Stupidly i thought i could just go in to each node on my cluster and change ip in the GUI and the jobs done. Whoops how wrong i was.

How do i get my cluster back up and running? do i need to do anything or do i have to start again, luckly i do have my vms backed up but would be a pain to reinstall both nodes.

Thanks for reading and any help.

Thanks
 
thanks found that the /etc/hosts was still on the old ips.

i did take a look at that link before posting. i got stuck where it says

Code: 
# put correct corosync config into local pmxcfs and corosync config dir (make sure to bump the 'config_version' inside the config file)
$ cp correct_corosync.conf /etc/pve/corosync.conf
$ cp correct_corosync.conf /etc/corosync/corosync.conf

where do i get the correct_corosync.conf file from? and waht does bump the config version inside the config file mean?

sorry for the basic questions.

Thanks for the reply.
 
If you still want to make the change you can update everything to the desired new ips of course.

In that thread the user already created a correct corosync config file.
But you can Just edit /etc/pve/corosync.conf and copy the changed file to /etc/corosync/corosync.conf.
In the totem section of that file is a value config_version which should be increase by 1 if you make changes so corosync knows it is a newer version.
 
Last edited:
thanks tried that but i got a few errors, when i ran pvecm status on node 2 i get the old IP still of 10.0.0.42 when it should be 192.168.1.42 i can get on to the GUI with 192.168.1.42

also when i run pvecm status on the first node i get this

1780575361888.png

node 1 top node 2 bottom.

any ideas what i am doing wrong?

Cheers for the help!
 
On pve01; did you copy the corosync config to /etc/pve/corosync.conf as well?

Looks like the corosync config is not yet updated on pve02. Did you restart the services or reboot after changing the config?

Do not try to fix two nodes at the same time. Do the second when you are sure the first node is configured properly.
 
  • Like
Reactions: mason64
Thank you for all your help! @daanw you have managed to help me fix my cluster and also the same time make me learn so much about clusters,

So i take it the corosync file needs to be th same on all nodes in /etc/pve and /etc/corosync also same goes for the authkey as i didnt have that on pve 1 so copied it to pve1 and it came back up

great help. love this forum
 
Glad to hear you got the cluster going again. You can mark the thread as solved then. :)

Once you have a working cluster and corosync detects a config_version higher than the previous version it will try to distribute the new corosync.conf to the other nodes. That is why you only have to update it on one node. (If at all, normally you shouldn't have to touch the config.)
 
Last edited:
You must log in or register to reply here.
Top Bottom
Back