boardy

I don't believe so, I built this VM quite some time as ago, so can't 100% remember, but I'm fairly sure all I did was provide the virtio driver ISO within the VM settings, I don't recall doing anything with virtio drivers directly from within Windows. The VM has been working for quite some time, several months until I noticed it recently stopped - not sure if UnRAID Upgrade may have broke it in some way.

As title suggests, I have a Windows VM which is mostly responsible for running backups on my network which it emails to say its completed. I realised I hadn't see any of these emails in a while and found the VM was shutdown. Turned it back on and connected to VNC and I then see the Windows boot up screen, then VNC says unable to connect and the VM is showing as shutdown again. At the bottom of the VM log I can see the following: I've tried a different browser as I've seen that could be the cause but no luck. Anyone else seen this or might know how to recover the VM.

Thanks Hoopster, That's interesting to know. I found an APC UPS on Amazon and asked the question there and APC responded themselves saying it only supports Windows, maybe they were just referring to the software for setting up. I think I'll get the APC one I found on Amazon then and hope that does the trick. Thanks for your help.

Hi everyone, I am looking at getting a UPS for my UnRAID server but having difficulty finding one. I'm looking for one that is compatible with UnRAID in the aim of shutting it down cleanly in the event of a power outage. I'm not too fussed about leaving it running for a long period of time on battery so nothing expensive, looking around £200ish if possible. The ones that I've found though have software which means the functionality for allowing safe shutdown is only Windows compatible - and presumably a Windows VM running on UnRAID wouldn't work. Can anyone recommend something?

Thanks. I can confirm that this does indeed appear to have worked. Thanks again.

Thanks for the reply, I missed the little helper text next to the dropdown, thinking about it I think I turned off the VMs (they were suspended automatically after the cache got filled). I assumed it moved the VMs as I when I was triggering the mover manually it kept stating it was skipping and not actually doing anything, and then shutdown the VMs and set this to No then triggered manually again and it started the move. I'm a tad confused now though whether the VMs are actually on the cache or on the HDDs, is there a way to determine where they are, or do I not need to worry, just move my 2nd disk images (where the big data is) onto a new share that uses cache=yes and the domains share to cache=prefer and run the mover again and it will just sort itself out?

I am new to UnRAID and I think I misunderstood how the caching works. What I was thinking is the VMs stored on cache, but will eventually move to the raid disks and then if the cache fills up the raid disks would be used by default. However I discovered this wasn't the case as the domains path (where the VMs are stored) is set to prefer cache so it stays on the cache drives and not the raided drives. Because I was migrating a VM from an existing platform, it had a lot of data so it filled my 512GB SSD cache drive and then the VMs just stopped because they ran out of space. So I was thinking initially, and wrongly, set to use the cache to no and then trigger the move so it moved all of the VMs to the raid 4TB mechanical HDD disks. I've realised my mistake, so my plan now is I will have the OS disks stay in the domains share so its cache is set to prefer so it stays on the cache. Then I plan to create another share for big data disks for the VMs so would have the share cache value set to yes which if I understand correctly, it will initially write to cache, but then get moved to the raid 4TB HDD disks where as prefer means stay on cache. Therefore assuming this plan is correct, I need to move my current VM disks back to the cache. Is that possible to move the VMs from the RAID HDD back to the cache or it will be a case of deleting the VMs and recreating them again using this new setup? Thanks

I wondered if it might be PSU but dismissed it because it only happens when the GPU is passed through to the VM (doesn't happen anymore either since I changed the surge protector extension the server was connected to).

I am new to UnRAID just installed this week and the process has been pretty straight forward, except for one thing which was working fine then just suddenly stopped. I suspect it might be a hardware fault but not 100% sure so wondering what everyone else's thoughts are. I've installed UnRaid on a Asus Z170 Pro Gaming Motherboard, Intel I7 CPU Skylake and 32GB DDR 4 of RAM with a pretty old Nvidia GeForce GTX 760. I successfully installed UnRAID and installed a Windows 10 Pro VM fine and then figured out to pass through the GPU to the VM hooked up to my TV and everything worked fine. Then I shut the VM down and the UnRAID server over night as there was no need to leave it on. The next morning booted it up and UnRAID booted fine, then turned the Windows 10 VM on, and UnRAID went offline (could no longer access Web GUI or ping IP address). I power cycled the server, and then during POST I got the message "Power Surge Detected - PSU triggered surge protection at last boot". I cleared the message thinking it was just a one time hiccup and then booted the VM again and then again lost access to UnRAID and power cycled and got the same message at POST about a power surge. I then changed the VM config back to VNC and then successfully booted the VM and set it on a static IP and turned off firewall temporary so I could ping the server and then shut back down and re-configured to use the GPU pass through again. Again UnRAID timed out and on power cycle got the power surge message again. I noticed the power surge protector I had used, the surge protection light had gone out so presumed the surge protector wasn't working so bought a new extension lead and re-tested, however now when I boot the Windows 10 VM UnRAID stays up and reported the VM as started, but it can't ping so its like its not booting for some reason when the GPU is passed through. I shutdown the VM and put back to VNC and then the VM boots successfully. The GPU at least works on the UnRAID boot (the non GUI haven't tried booting it with the GUI) so wondering if maybe the GPU has developed a fault where it can copy with displaying the terminal output from UnRAID on boot but anything slightly more demanding such as the Windows 10 GUI it fails. I hope its not a GPU failure as now would be a really bad time to by a new GPU. Anyone else seen this before or got any advice to debug what the issue is. I unfortunately don't have another PC I can use to plugin the GPU to a normal PC and see if it behaves any differently. Thanks Chris