noVNC keeps disconnecting every minute or so

[naylo44]

Hi everyone! So I've just reinstalled my Proxmox cluster and I'm encountering an annoying issue with the noVNC console for both VMs and the Hosts. It keeps disconnecting every minute for 5 to 10 seconds or so. I will attach a sample from the node's Syslog.

Nov 20 22:43:24 pve3-g5 pvedaemon[48920]: <root@pam> end task UPID:pve3-g5:0000E61B:0008F429:6199BCB1:vncproxy:211:root@pam: OK
Nov 20 22:44:01 pve3-g5 pvedaemon[68300]: starting vnc proxy UPID:pve3-g5:00010ACC:000A7161:6199C081:vncproxy:211:root@pam:
Nov 20 22:44:01 pve3-g5 pvedaemon[48920]: <root@pam> starting task UPID:pve3-g5:00010ACC:000A7161:6199C081:vncproxy:211:root@pam:
Nov 20 22:44:22 pve3-g5 pvedaemon[1982]: <root@pam> end task UPID:pve3-g5:0000E820:00090A9A:6199BCEB:vncproxy:211:root@pam: OK
Nov 20 22:44:54 pve3-g5 pvedaemon[51026]: <root@pam> starting task UPID:pve3-g5:00010CD4:000A85EC:6199C0B6:vncproxy:211:root@pam:
Nov 20 22:44:54 pve3-g5 pvedaemon[68820]: starting vnc proxy UPID:pve3-g5:00010CD4:000A85EC:6199C0B6:vncproxy:211:root@pam:
Nov 20 22:45:14 pve3-g5 pvedaemon[48920]: <root@pam> end task UPID:pve3-g5:0000EA57:0009216C:6199BD25:vncproxy:211:root@pam: OK
Nov 20 22:45:47 pve3-g5 pvedaemon[69363]: starting vnc proxy UPID:pve3-g5:00010EF3:000A9A93:6199C0EB:vncproxy:211:root@pam:
Nov 20 22:45:47 pve3-g5 pvedaemon[51026]: <root@pam> starting task UPID:pve3-g5:00010EF3:000A9A93:6199C0EB:vncproxy:211:root@pam:
Nov 20 22:46:14 pve3-g5 pvedaemon[1982]: <root@pam> end task UPID:pve3-g5:0000EC50:00093586:6199BD59:vncproxy:211:root@pam: OK
Nov 20 22:46:44 pve3-g5 pvedaemon[51026]: <root@pam> starting task UPID:pve3-g5:00011171:000AB0FB:6199C124:vncproxy:211:root@pam:
Nov 20 22:46:44 pve3-g5 pvedaemon[70001]: starting vnc proxy UPID:pve3-g5:00011171:000AB0FB:6199C124:vncproxy:211:root@pam:
Nov 20 22:47:09 pve3-g5 pvedaemon[51026]: <root@pam> end task UPID:pve3-g5:0000EE6F:00094D13:6199BD95:vncproxy:211:root@pam: OK
Nov 20 22:47:37 pve3-g5 pvedaemon[70571]: starting vnc proxy UPID:pve3-g5:000113AB:000AC56B:6199C159:vncproxy:211:root@pam:
Nov 20 22:47:37 pve3-g5 pvedaemon[51026]: <root@pam> starting task UPID:pve3-g5:000113AB:000AC56B:6199C159:vncproxy:211:root@pam:
Nov 20 22:48:02 pve3-g5 pvedaemon[48920]: <root@pam> end task UPID:pve3-g5:0000F06B:000962AB:6199BDCC:vncproxy:211:root@pam: OK
Nov 20 22:48:04 pve3-g5 pvedaemon[1982]: worker exit
Nov 20 22:48:04 pve3-g5 pvedaemon[1980]: worker 1982 finished
Nov 20 22:48:04 pve3-g5 pvedaemon[1980]: starting 1 worker(s)
Nov 20 22:48:04 pve3-g5 pvedaemon[1980]: worker 70848 started
Nov 20 22:48:28 pve3-g5 pvedaemon[51026]: <root@pam> starting task UPID:pve3-g5:000115C2:000AD96B:6199C18C:vncproxy:211:root@pam:
Nov 20 22:48:28 pve3-g5 pvedaemon[71106]: starting vnc proxy UPID:pve3-g5:000115C2:000AD96B:6199C18C:vncproxy:211:root@pam:
Nov 20 22:49:26 pve3-g5 pvedaemon[70848]: <root@pam> starting task UPID:pve3-g5:00011817:000AF056:6199C1C6:vncproxy:211:root@pam:
Nov 20 22:49:26 pve3-g5 pvedaemon[71703]: starting vnc proxy UPID:pve3-g5:00011817:000AF056:6199C1C6:vncproxy:211:root@pam:
Nov 20 22:49:52 pve3-g5 pvedaemon[48920]: <root@pam> end task UPID:pve3-g5:0000F49D:00098DBB:6199BE3B:vncproxy:211:root@pam: OK

This happens when having a noVNC window opened on a VM residing on my node pve3-g5. It happens regardless of the browser I use (issue happens with Chrome, Edge and Firefox).

The VM itself runs fine, I was able to do everything I needed over SSH and even installed a linux VM with a Plasma DE + NoMachines and I was able to use it normally, without the disconnects.

Any idea where to look for to fix this issue?

Thanks

 

[Moayad]

Hi,

What is your PVE version pveversion?
Have you tried on a browser without plugins/private browser?

 

[W4RCL0UD]

For anyone else running into this - I was having the same issue as OP and resolved it by disabling the windows firewall - I haven't been able to nail down exactly what is causing the disconnects but can confirm it's running perfect after disabling the firewall... would love to see if anyone else has created a firewall rule to resolve this versus just disabling the entire profile.

 

[naylo44]

I forgot this post.

This issue seems to stem from the fact that the machine I used to access the web ui from is in 192.168.10.x/24 and I was accessing the proxmox server residing on another subnet (192.168.20.x/24). I added another vlan on the proxmox machines in the 192.168.10.x vlan and everything works now. No clue why the traversal of Vlan broke stuff.

 

[calperin]

So, this is the same issue that the TrueNAS Management? Exactly the same sympthom.

 

[alexraio]

Hello everybody,
I have exactly the same problem. If I use the Web UI from a machine In the same network, than it works everything fine. If I reach the Web UI from a machine in a different network than noVNC keeps disconnecting. Very annoying. Have anybody an idea what could be the problem?
Thank you in advance.

 

[alexraio]

Would be really appreciated if the Proxmox development team could analyze this behavior. In an enterprise environment multi network configuration is a standard.

 

[gustavoppires]

Hello everybody,
I have exactly the same problem. If I use the Web UI from a machine In the same network, than it works everything fine. If I reach the Web UI from a machine in a different network than noVNC keeps disconnecting. Very annoying. Have anybody an idea what could be the problem?
Thank you in advance.

I have also the same issue. I can't change subnets as I'm using my work pc to access PVE and noVNC. Very annoying.