Hey all, came across a weird issue that I can't seem to sort out. Randomly, after shutting down one of my VMs, the VM Manager portion of the WebGUI will just go completely blank. Libvirt seems to still be running and I'm not seeing any errors to point to what could be going on. Libvirt logs don't seem to show anything out of the ordinary.

 

2025-02-21 00:44:32.501+0000: 12000: info : libvirt version: 10.7.0
2025-02-21 00:44:32.501+0000: 12000: info : hostname: Tower
2025-02-21 00:44:32.501+0000: 12000: warning : qemuDomainObjTaintMsg:7164 : Domain id=1 name='Bazzite' uuid=029f5f26-639a-02f9-506b-c313bba7f4a7 is tainted: high-privileges
2025-02-23 01:14:16.370+0000: 12000: warning : qemuDomainObjTaintMsg:7164 : Domain id=2 name='Bazzite' uuid=029f5f26-639a-02f9-506b-c313bba7f4a7 is tainted: high-privileges

 

Here's what the VM Manager looks like.

 

 

If I attempt to stop the libvirt service and start it back up, I get a message that libvirt failed to start. With this message in the system logs.

Feb 23 18:35:23 Tower emhttpd: shcmd (260): /usr/local/sbin/mount_image '/mnt/user/system/libvirt/libvirt.img' /etc/libvirt 1
Feb 23 18:35:23 Tower root: '/mnt/user/system/libvirt/libvirt.img' is in-use, cannot mount

If I reboot the Unraid server entirely, libvirt starts back up and the VM Manager WebGUI shows back up.

 

Any ideas where I should start looking to track down what is going on here?

10 hours ago, tool666schism said:

If I attempt to stop the libvirt service and start it back up, I get a message that libvirt failed to start.

This can happen if containers are running, stop all containers, or the docker service, and try again.

5 hours ago, JorgeB said:

This can happen if containers are running, stop all containers, or the docker service, and try again.

I ended up rebooting the server to get the gui back loaded, but next time it fails on me I'll give that a try and see if that works.

 

Any clue why the VM Manager is going blank in the first place? I haven't seen anyone else reporting similar issues so not sure if this is a bug or a configuration issue on my end.

Is it just the VM manager? Post new diags after it happens again.

Yea it's just the VM Manager. Of course I can't get it to replicate it now but as soon as I can I'll post diags. Thanks for the help.

I'm experiencing the same behaviour with VM manager GUI. The VM list is also blank on the dashboard.

 

The VM is still running according to virsh command in the terminal.

 

The running VM is Windows 11 and is a dual boot PC with unraid.

 

The only solution I have found is a reboot, which isn't ideal.

 

Nick

tower-diagnostics-20250224-2039.zip

Glad to see i'm not alone with this issue. The frustrating thing is I can't seem to replicate it consistently. It happened 3 times yesterday. Then I create this post and it hasn't happened again since. I've been starting and stopping VM's all day and it hasn't broken yet.

 

My VMs are both linux, bazzite and pop_os.

Unraid connect displays the correct information, so this looks like an unraid UI issue on some sort.

 

 

I've updated to 7.01 and launched my W11 VM. I will update this thread if the problem re-occurs.

Ok, so after the VM has been running for just under a day I spotted that the problem has reoccurred.

 

Updated diagnostics attached

 

Hope someone has some ideas of what to try to resolve the issue of confirm the bug.

thanks

Nick

 

tower-diagnostics-20250227-1735.zip

As a first step, I would recommend booting in safe mode to rule out any plugin issues.

ok, I have just launched the vm after a safe mode boot and will report back.

I can confirm that the problem also occurs when booting in safe mode, so looking for next steps please to progress this further.

 

Diagnostics from when the issued occurred in safe mode attached.

 

 

tower-diagnostics-20250301-1058.zip

There are PCIe errors spamming the log, making it very difficult to look for anything else, see if this helps, then post new diags after it happens again.

https://forums.unraid.net/topic/118286-nvme-drives-throwing-errors-filling-logs-instantly-how-to-resolve/?do=findComment&comment=1165009

 

Thanks JorgeB. I've added the boot entry from the linked post, rebooted and launched the VM. I will post updated diags as requested.

Nick

Hi JorgeB, please find attached updated diagnostics as requested.

Regards

Nick

tower-diagnostics-20250303-0208.zip

Nothing obvious that I can see, though there are a lot of USB disconnect messages logged, any idea what those are about?

I've not noticed any USB device issues when working in the Windows 11 VM, so no idea what the disconnect messages are about - shame it doesn't specify the device.

 

From recollection I think this only happens when the VM is running, but I will retest by rebooting and not running the VM to check that this doesn't happen when the VM is stopped.

Nick

 

 

Hi JorgeB, here's an update.

 

- With the VM in a stopped state the problem did not reoccur

- Several hours after then starting the VM, the GUI went blank again. I was lucky enough to spot within 5mins of when this happened, checked the logs and there were no entries at all at this time

- I also tested with the VM usage statistics disabled, but the problem still re-occurred

- If I leave the VM page open on a second device (and don't refresh it), then I can still interact with the GUI (after it is blank on the other device), but the usage stats % do not change, so it is frozen, but I can still select start, stop etc. (although if I select edit, the page does not load beyond the logo at the top)

 

Any other ideas?

Nick

I would try creating a new VM pointing to the same vdisk, then retest.

Hi JorgeB,

 

So, as requested I ran a new W11 VM for a couple of days and the problem did not reoccur. I then also started my main VM and the problem has reoccurred.

 

I should point out that my main VM is using my bare metal install of W11 which is on a separate NVME drive with the controller passed through.

 

Nick

Did you try creating a new VM but pass-through the same NVMe device?

Hi JorgeB,

 

No, so I have done that now and can confirm the problem has re-occurred almost immediately.

 

Nick

Edited March 8, 20251 yr by stampie

OK, that suggests the issue may be related to the NVMe device, or it being passed through, assuming no more difference between the first new VM and the one using the device.

16 hours ago, JorgeB said:

OK, that suggests the issue may be related to the NVMe device, or it being passed through, assuming no more difference between the first new VM and the one using the device.

I can confirm that I was using an NVMe drive that was being passed through to the VM as well. Seems like that may very well be the culprit. I have been using a brand new VM in the meantime to try and figure out why this was happening but I had not yet passed the NVMe through to it and the issue hadn't re-occured.