[SOLVED] Failed to connect to display of VM - Cluster

[Iamfrank]

Hello,

I have noticed some strange behaviour and don't know how to deal with my cluster.

3 Proxmox VE 8.0 servers (updated from 7.x) are working as one cluster.
The update was straightforward and I don't think it has anything to do with the update, I just didn't notice it before.

Server #1, #2, #3 contain VMs - server #3 got a fresh one after the update for testing.

Cluster works over 2 Networks as recommended. I can see and access any server over any control panel.

During the total update of all servers, I moved live VMs form #1 to #2 and back (awesome) no problem.

VMs work like a charm.

But

Logging into #1 - control panel, I can open the displays "consoles" of the VMs for #1 and #2, but not the VMs of #3.
Logging into #2 - control panel, I can only open the display for VMs of #2.
Logging into #3 - control panel, I can only open the display for VMs of #3.

It is obviously a cluster issue.

Error massages are "Failed to connect to server" (VNC) or "unable to connect to graphic server" (spice).
I have restarted my coresponding computer, webbrowser and refreshed the webpages and used a new webbrowser to be sure that there is not a cookie issue.

I suspect it's a problem with the security keys, but I have no idea how to update my keys if I had to.

Any help is very welcome.

 

[Philipp Hufnagl]

Can you try to connect from #1 and afterwards post the syslog of #3 ?

 

[Iamfrank]

Hallo Philipp,

Being on #1 and
looking on the syslog #3
Starting the #3 VM with vnc failing to connect.

LOG:

Jul 06 15:25:55 Proxmox03 kernel: device tap301i0 entered promiscuous mode
Jul 06 15:25:55 Proxmox03 kernel: vmbr0: port 2(fwpr301p0) entered blocking state
Jul 06 15:25:55 Proxmox03 kernel: vmbr0: port 2(fwpr301p0) entered disabled state
Jul 06 15:25:55 Proxmox03 kernel: device fwpr301p0 entered promiscuous mode
Jul 06 15:25:55 Proxmox03 kernel: vmbr0: port 2(fwpr301p0) entered blocking state
Jul 06 15:25:55 Proxmox03 kernel: vmbr0: port 2(fwpr301p0) entered forwarding state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 1(fwln301i0) entered blocking state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 1(fwln301i0) entered disabled state
Jul 06 15:25:55 Proxmox03 kernel: device fwln301i0 entered promiscuous mode
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 1(fwln301i0) entered blocking state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 1(fwln301i0) entered forwarding state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 2(tap301i0) entered blocking state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 2(tap301i0) entered disabled state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 2(tap301i0) entered blocking state
Jul 06 15:25:55 Proxmox03 kernel: fwbr301i0: port 2(tap301i0) entered forwarding state
Jul 06 15:25:55 Proxmox03 pvedaemon[3268]: <root@pam> end task UPID:Proxmox03:00010BC5:00092F0E:64A6C0E2:qmstart:301:root@pam: OK
Jul 06 15:26:10 Proxmox03 pveproxy[65826]: Clearing outdated entries from certificate cache
Jul 06 15:26:48 Proxmox03 pmxcfs[3089]: [status] notice: received log
Jul 06 15:27:37 Proxmox03 pmxcfs[3089]: [status] notice: received log
Jul 06 15:27:37 Proxmox03 sshd[71044]: Connection closed by 192.168.188.229 port 54166 [preauth]
Jul 06 15:27:37 Proxmox03 pmxcfs[3089]: [status] notice: received log

---

 

[Philipp Hufnagl]

There might be a cert issue for #3

Are you using self signed certificates?

If so try running

pvecm updatecerts --force

on all the VMs

If you user your own certificate (like letsencrypt) check if your certificate has been correctly installed/deployed on #3

 

[Iamfrank]

Thank you for the tip - it worked now!!

I have tried this several times with

pvecm updatecerts -f

with no success.

But now I have not only

pvecm updatecerts --force

applied to all my servers #1-#3, but also rebooted everything and closed the browser!

Now the connections are working again as before.

Thank you very much for your help, I really appreciate it.