Belcraig

whenever the card is in. it dosnt matter if the VM is running will set up a syslog and see what that posts

Hi All, Not entirely sure what could be going wrong recently with my unraid server. a little while ago i started to have issues with the gaming VM and unraid not releasing the GPU. this was fixed by removing the rom file. However my server keep on restarting and as part of the process, the settings on my HBA card are reset. Running the GPU in a different computer runs fine and stressing it doesnt' produce errors to I'm therefore fairly certain that the GPU is good. The HBA has three drives attached and has now been online in unraid for 3 days with no error. however re-introducing the GPU back into the setup causes the server to restart after a few hours. Moving the HBA to different PCIe slots does not cause it to fail. Any one got ideas what might cause this.or something else i could test, i've a feeling something in an unraid update/or myself altered a setting somewhere that upset Unraid and so started the GPU failing to be passed through.and trying to fix that has now caused this. Its a duel power supply with wattage's being drawn quite consistent across both. I've also tried separately incase one of the supplies are going. but i get the same issue so fairly sure its not a power supply issue either. thanks for any suggestions.

HI all, In case anyone else finds themselves in a similar problem it seems that the ROM was incorrect... having it now not in the primary slot means i do not need the ROM. Not sure if an unraid update made the ROM invalid.. Regards

ok I've moved the PCIE cards around so that the nvidea gt710 sits in the primary slot on the mother board, and moved the rx 570 to a different slot. I've acknowledged the PCI changes in devices settings and bound the IOMMU group to VIFO at Boot (the rx570). restarted the server and still no images on the screens from the VM. Anyone else able to throw an idea? I am getting this in the logs but if i bound both the GPU and the sound at boot how can i stop it picking this up. 2026-02-17T15:40:21.402594Z qemu-system-x86_64: vfio: Cannot reset device 0000:82:00.1, no available reset mechanism. 2026-02-17T15:40:21.510705Z qemu-system-x86_64: vfio: Cannot reset device 0000:82:00.1, no available reset mechanism. 2026-02-17 15:40:28.437+0000: Domain id=1 is tainted: custom-monitor usb_generic_handle_packet: ctrl buffer too small (20480 > 4096) thanks Craig

the normal operation - what has been working for the last several years is unraid will boot into GUI, if i then start a VM the VM takes the graphics card, Unraid never gets it back from this without a reboot of the server. However, this has stopped working. I dont necessarily need the GUI i can go to a different room and access if there (if the VM has an issue) -its mroe convienance. However, I've tried to bind the GPU to VIFO at boot in the tools --> system devices but this is not working, on the screen i get the bios and the loading, then the screen stops but it dosn't pass through to the VM.

HI all, last few months i've been having an issue with my VM on unriad, i've used co-pilot a few times to help correct the issue, but it keeps comenig back so i thought i'd try here for help. For several years I've been using one graphics card in my server that on boot loads into the unraid GUI, but when i launch my main VM it disconnects from the server and launches the vm using that card. I've had no problems with this all all, until the last few months, i'm not sure when it first happened or if it was related to an update. I've had a fail drive aroudn the same time i first noticed this, but no other harware changes, and i havn't tinkered with the setting on the system. I've tried to bind the GPU to the VIFO on boot but i still get the same issues. any one got any clues or ideas that i can try? i have a second GPU in the system but i cannot alter the placement in the PCIE slots as they will not fit. thanks for your help in advance. server-diagnostics-20260210-2207.zip

root@Server:~# btrfs balance start -dconvert=raid1 -mconvert=raid1 /mnt/cache ERROR: error during balancing '/mnt/cache': Read-only file system root@Server:~# dmesg | tail [42390.025192] cdc_acm 3-12:1.0: ttyACM0: USB ACM device [42393.486851] usb 3-12: USB disconnect, device number 35 [42393.790122] usb 3-12: new full-speed USB device number 36 using xhci_hcd [42393.919642] cdc_acm 3-12:1.0: ttyACM0: USB ACM device [45993.120095] usb 3-12: USB disconnect, device number 36 [45993.370983] usb 3-12: new full-speed USB device number 37 using xhci_hcd [45993.500009] cdc_acm 3-12:1.0: ttyACM0: USB ACM device [45996.961219] usb 3-12: USB disconnect, device number 37 [45997.271974] usb 3-12: new full-speed USB device number 38 using xhci_hcd [45997.401033] cdc_acm 3-12:1.0: ttyACM0: USB ACM device

how to do this please...

I stopped the array, it now shows three slots? when i select one of the disks in place of the missing device i get cache too many wrong or missing devices.

server-diagnostics-20250909-1715.zip I've done as instructed..

opps didnt realsie the difference. Label: none uuid: 01cc169c-ab3e-4f2b-86e3-5382b85f07e7 Total devices 3 FS bytes used 471.12GiB devid 2 size 931.51GiB used 474.36GiB path /dev/sdc1 devid 4 size 238.47GiB used 238.47GiB path /dev/sdh1 *** Some devices missing Label: none uuid: 8f7ab551-6fc8-455a-b371-6121ae81d696 Total devices 1 FS bytes used 144.00KiB devid 1 size 1.82TiB used 2.02GiB path /dev/sdf1

Ok so i apologize for being inpatient... i did the above formatted the drive (btrfs) added it to the pool and re-started the server. No access to GUI. Ah i thought i know how to fix this... deleted config.cfg from pools. restarted the server. Access back to gui Yeah!, re-created the pool and added the three drives, restarted the server... Unmountable :( server-diagnostics-20250908-2002.zip

followed the above and, re-added the drive back to the array and still get unmountable. server-diagnostics-20250907-1738.zip

thankyou, that works, how can i add the drive back into the pool then - or is that a lost drive...

server retarted and pool mounted with the two disks as before server-diagnostics-20250905-1925.zip