Hello Unraid Community

Since updating to Unraid 7.1, I’ve been experiencing a strange issue: whenever I start any of my VMs, the Web UI and Docker services crash. I updated to version 7.2 today hoping it would resolve the problem, but unfortunately, the issue persists.

Once a VM (either one of the Windows 11 VMs or the HomeAssistant VM) is started, after 2–5 seconds, the Web GUI, Docker containers, and network shares become completely inaccessible. However, the Windows VMs themselves remain accessible via RDP without any problems. If I avoid starting any VMs, the server runs normally, shares and Docker containers are fully functional.

When I try to boot Unraid into GUI Mode or Safe GUI Mode, the GUI never appears. I’m forced to perform a hard reset to get the system running again. The normal command line interface works without any problems. Could this be related to the VM issue?

Has anyone else encountered this or have suggestions on where to start troubleshooting? As I have been running the server for the past year without starting any VMs just to avoid the problem. I’ve attached diagnostics in case they help.

Thanks in advance for your help!

Also, on another note, because of this, I’m considering starting fresh by installing a stock Unraid image onto the USB drive and rebuilding everything from scratch. Could someone point me to a guide on how to perform a clean install of Unraid --> preserving only the Array (disks with encryption) shares and their data?

urbiverse-diagnostics-20251108-1512.zip

Edited November 8, 2025Nov 8 by urbi

Just to confirm, are the diags after a crash? Also, if you create a new test VM, do you see the same issue?

Have you followed the advice here for the Docker side?

https://docs.unraid.net/unraid-os/troubleshooting/common-issues/docker-troubleshooting/

42 minutes ago, JorgeB said:

Just to confirm, are the diags after a crash? Also, if you create a new test VM, do you see the same issue?

Hallo JorgeB

Yes, the diagnostics were taken after the WebUI went down (using the diagnostics command in the terminal). After that, I issued the reboot command to bring the WebUI back online.

I have created a new Windows 11 VM, and the issue with the WebUI is no longer present on the new instance. Attached below is the XML configuration of the problematic VMs for your reference.

I am considering recreating the VM from a fresh template. Could you advise which values I should copy from the old template to the new one (UUID, MAC address, etc.)? Alternatively, if you spot any problematic settings in the XML that might have caused the issue, I would appreciate your insights.

Thanks in advance!

@leejbarker I am not using docker.img, but rather the Docker directory. Does the recreation process also apply when using the directory method?

Home Assistant VM.xmlWindows VM.xml

I'm going to say probably not, but I can't be 100% certain.

Don't see anything obvious in the syslog, but if a new VM works, it may be worth reading them, I would think that keeping the UUID and pointing to the existing vdisks would be enough, but I recommend trying it on a cloned vdisk first and keeping the old XMLs intact for now.

Hallo JorgeB

I have tried several approaches, but none have been successful so far. The same behavior occurs with both the Home Assistant and Windows virtual machines:

- I recreated the virtual machine and copied over the UUID and MAC address from the original. Upon starting it, the WebUI went down.

- I created a new virtual machine without copying any settings. I only pointed it to the existing virtual disk. The WebUI still crashed.

- I renamed the libvirt.img file and allowed Unraid to generate a fresh one. Then I created a new virtual machine and linked it to the same disk image. Unfortunately, the WebUI crashed again.

I have attached two diagnostics files:

- One taken after creating a new virtual machine and pointing it to the existing disk.

urbiverse-diagnostics-20251109-1443-new_vm.zip

- Another taken after generating a new libvirt.img and creating a fresh virtual machine linked to the same disk.

urbiverse-diagnostics-20251109-1458-new_libvirt_vm.zip

That suggests to me there's something in the actual vdisks causing the problem if a new VM doesn't have issues.

Thanks, JorgeB

The only mystery that remains is why all three virtual machines became corrupted at the same time—two were running and one was shut down. Unfortunately, I have come to the realization that those virtual machines are a lost cause.

Over the weekend, I was contemplating whether I should rebuild the entire Unraid server from scratch, preserving only the shares. Everything else I can set up anew. Interestingly, I also discovered today that someone else is considering starting their Unraid journey fresh, and I believe this might be the best path forward for me as well.

The matter of the WebUI crashing is now closed for me.

Nevertheless, I would like to invite you and anyone else who may come across this post, to share any input or suggestions on how to start with Unraid from scratch while preserving existing data.

Thank you in advance for any pointers or recommendations or is there an official guide available that explains how to do this properly?