Proxmox node status is showing as unknown

admlvntv

New Member
Apr 1, 2026
2
0
1
Hello,
I have a freshly installed Proxmox server that was just added to a newly made cluster, and now has its status shown as unknown and is not able to create VMs. However, all of its system information is fully functional, and I am able to SSH into it. I have tried checking pvestatd and nothing out of the ordinary appears. Running journalctl -eu pvestatd shows nothing special. Any advice?
1775079590940.png1775079646976.png

1775079761470.png
 
Hi admlvntv,

Welcome to the forums!

These are screenshots taken of the pve02-side of things?

What is in both servers' /etc/hosts?
 
  • journalctl -b (upload as file here)
  • cat /etc/pve/corosync.conf on both nodes
  • pveversion -v
  • systemctl status corosync
  • cat /etc/network/interfaces
P.S.: Welcome to the Forum!
 
Hi all, thanks for the quick response and warm welcomes (:


Hi admlvntv,

Welcome to the forums!

These are screenshots taken of the pve02-side of things?

What is in both servers' /etc/hosts?

The screenshots are taken for pve02, but using the webui of bioumass, as the pve02 web ui went down; unsure if that is expected behavior. The hosts file has nothing out of the ordinary, just the defaults.
Screenshot 2026-04-03 at 6.26.10 PM.pngScreenshot 2026-04-03 at 6.26.20 PM.png

  • journalctl -b (upload as file here)
  • cat /etc/pve/corosync.conf on both nodes
  • pveversion -v
  • systemctl status corosync
  • cat /etc/network/interfaces
P.S.: Welcome to the Forum!
I currently only have shell access to pve02, so I attached those in the meantime. I'll send a follow up post when I get the corosync.conf from bioumass. Note that bioumass is running 8.4.1 and pve02 is running 9.1.6; we are planning a migration.
Screenshot 2026-04-03 at 6.32.08 PM.pngScreenshot 2026-04-03 at 6.32.25 PM.png
 

Attachments

Last edited:
Hi admlvntv,

I currently only have shell access to pve02

"quoting" screenshots is not terribly useful, and screenshots in general are not that useful most of the time:
  • The forum search engine does not have access to the text in the screenshots, which does not help subsequent visitors to solve their problems
  • The thumbnails are often too small to read, while the full image is often larger than a (non full screen) browser window
  • Copy/pasting a certain error from a screenshot is not possible
Yes, it means that whoever needs help at that moment has to take the step of copying text and using the </> code-button to layout the text as code, instead of only press Printscreen and Ctrl-V.

So far my rant ;-)

The hosts file has nothing out of the ordinary, just the defaults.

It seems that one of the hosts has a public IPv4, while the other has a non-routable IPv4.
Neither of them has a SLAAC or publick IPv6.

My guess is that the hosts are in the same physical network and can reach each other by accident (hence you can manage pve02), but the logic that sets the question mark may not be aware of that.

What about
Code:
ip a
on both hosts, and
Code:
ip route
?

Edit/PS: having had a look at the attachments: I see that bioumass has a private IPv4 as well. I'd add it to the hosts file.
 
Last edited:
You have configured your corosync to use the internal ip addresses 172.24.129.85 (biomass) and 172.24.124.27 (pve02) but your machine "biomass" has the ip not configured. Thats why your corosync is not working.

Did you change the IP of "biomass" after setting up the cluster?
 
  • Like
Reactions: Johannes S