New Host Giving Issues

[RudiL]

Hello all,

I'm new to Proxmox. I have 2 new nodes that i wanted to add to an existing cluster.
Node 1 - added all good.
Node 2 - not so much.

It started with "Failed to connect to server" when trying to connect to the console of the VM running the host.
i then tried to update the certificates - located in etc/pve and etc/pve/nodes respectively with the same certs(self signed internal) that i used on Node1 where all is aok.

Not sure if messed up the process but now its really broken.
When i try to view the system information for the Node - anything really no info comes back and it just eventually times out with a communication error 596.
How do i fix this?

last resort i guess is to remove the host from the cluster and rebuild it... but if i do that i would like to add it with the same name and IP. I read somewhere that is not advised.
Thanks for the help.

 

[RudiL]

Morning all,
Anyone with ideas?

 

[Matthias.]

Please provide some information, especially the output of
* pveversion
* pvecm status

Can you ping the other node?
Are there any messages in the syslog?

 

[RudiL]

Please provide some information, especially the output of
* pveversion
* pvecm status

Can you ping the other node?
Are there any messages in the syslog?

Hello Matthias: detail

 

[RudiL]

syslog does not provide me any information - totally blank.
i can still connect to the host via Putty and from there i dont seem limited to any functions.
From within the cluster however the server is of no use. test VM from there is down and will not run or migrate to another node.
I have a working cluster with the exception of this new node that does not have any workloads yet so proceeding with caution since i'm not an proxmox expert.

 

[Matthias.]

So when you cat /var/log/syslog or journalct -b, you get no output at all? That's rather strange.
What's the output of ls -l /var/log/syslog? Anything unusual in the logs of the other nodes?

Before you added the node, you could open the VM console without problems?
Can you still access the VM console via a different node?
Do you connect to the node itself, or to a different one? Does the problem exist on both?

 

[RudiL]

Output:ls -l /var/log/syslog:


cat /var/log/syslog = Output.txt

Before you added the node, you could open the VM console without problems? No but the VM ran no problem and could RDP to it.
Can you still access the VM console via a different node? Yes if i migrated it to another node i could.
Do you connect to the node itself, or to a different one? i connect to my first node in the cluster. From there i see all my nodes. When expanding node 6 i'm not able to do so. Before i could but then in attempt to solve the VM console issue i updated the certs again. And since then it have this issue.

 

[Matthias.]

Okay, did you execute pvecm updatecerts --force?
What's the output of systemctl status pve-cluster?

 

[RudiL]

Okay, did you execute pvecm updatecerts? No i did not - is it safe to execute this on my faulty node?
What's the output of systemctl status pve-cluster?

 

[Matthias.]

pvecm updatecerts --force
It should be safe, but will remove your existing keys and I believe the directories in which they are. So if you got some important data stored there, backup it first : )

 

[RudiL]

pvecm updatecerts --force
It should be safe, but will remove your existing keys and I believe the directories in which they are. So if you got some important data stored there, backup it first : )

ok so i run this on my 1 faulty node - will this have any impact on my other working nodes? I dont want to break a working cluster

 

[Matthias.]

The other nodes should be fine. The command changes the ssh keys of the current node.

BTW, to quote the docs:

Do not replace or manually modify the automatically generated node certificate files in /etc/pve/local/pve-ssl.pem and /etc/pve/local/pve-ssl.key or the cluster CA files in /etc/pve/pve-root-ca.pem and /etc/pve/priv/pve-root-ca.key.

 

[RudiL]

Thanks Matthias.
so i ran the command: i dont think it quite worked.

 

[Matthias.]

Okay, so it seems you changed your root certificate? As you can see in the error message, it say certificate and private key don't match.
You can maybe fix this by
1. taking a backup of all your certs (at least the /etc/pve/ folder!)
2. removing the root certificate at /etc/pve/priv/pve-root-ca.key
3. executing pvecm updatecerts --force