lelejau

Hello, I am looking for help investigating what appears to be my Unraid USB boot device dropping offline while Unraid was running as a VM on Proxmox. This setup had been stable for several months and no hardware or VM configuration changes had been made recently. Environment Unraid version: 7.3.2 Hypervisor: Proxmox VE 9.2.3 Proxmox kernel: 7.0.12-1-pve VM ID: 100 VM name: Unraid VM configuration visible in Proxmox: - BIOS: SeaBIOS - Machine: i440fx - RAM: approximately 20.94 GiB - CPU: 16 cores, type host - Boot helper ISO: plopkexec.iso - Network: VirtIO - USB boot device passthrough: usb0: host=0781:5567 - PCI passthrough devices: 0000:05:00 0000:00:1f.2 0000:02:00.0 The Unraid boot device is: Manufacturer: SanDisk Product: Cruzer Blade Vendor/Product ID: 0781:5567 Capacity: approximately 15.4 GB / 14.3 GiB The physical Proxmox host is connected to an APC UPS. The UPS is attached to the Proxmox host over USB and remained visible: Vendor/Product ID: 051d:0002 Product: APC Back-UPS BX 1500BI-BR Firmware: V586 There was no known power outage and the Proxmox host itself did not reboot. Today, at approximately 02:49BRT I tried to SMB into my shared folder. Got an error. I then opened the Proxmox console for the Unraid VM. Console login as root still worked, so I executed: reboot. The console displayed: Broadcast message from root@tower (tty1) (Wed Jul 15 02:49:34 2026): The system is going down for reboot NOW! INIT: Sending processes configured via /etc/inittab the TERM signal /etc/rc.d/rc.6: line 22: /etc/rc.d/rc.6: Input/output error INIT: no more processes left in this runlevel The VM then remained stuck indefinitely. It did not reboot or power off. From the Proxmox host I attempted a graceful shutdown: qm shutdown 100 --timeout 300 The result was: VM quit/powerdown failed - got timeout I then attempted: qm stop 100 The result was: trying to acquire lock... can't lock file '/var/lock/qemu-server/lock-100.conf' - got timeout At that point the guest appeared to be completely stuck in the shutdown process. I ultimately had to force-reset/stop the VM from Proxmox. I ran usb-devices on the Proxmox host. At that time it listed the USB root hubs and the APC UPS, but the SanDisk Cruzer Blade was not present. The relevant USB controllers on the host were: Linux 7.0.12-1-pve xhci_hcd Controller PCI address: 0000:00:14.0 I then ran: dmesg -w I physically removed and reinserted the SanDisk USB flash drive. Immediately after reinserting it, the Proxmox kernel logged: [861490.173243] usb 3-2: new high-speed USB device number 10 using xhci_hcd [861490.297590] usb 3-2: New USB device found, idVendor=0781, idProduct=5567, bcdDevice=1.00 [861490.297595] usb 3-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [861490.297597] usb 3-2: Product: Cruzer Blade [861490.297598] usb 3-2: Manufacturer: SanDisk [861490.297600] usb 3-2: SerialNumber: 04014718102423223929 [861490.298338] usb-storage 3-2:1.0: USB Mass Storage device detected [861490.298438] scsi host0: usb-storage 3-2:1.0 [861491.331787] scsi 0:0:0:0: Direct-Access SanDisk Cruzer Blade 1.00 PQ: 0 ANSI: 6 [861491.333593] sd 0:0:0:0: Attached scsi generic sg0 type 0 [861491.346252] sd 0:0:0:0: [sda] 30081024 512-byte logical blocks: (15.4 GB/14.3 GiB) [861491.347232] sd 0:0:0:0: [sda] Write Protect is off [861491.347235] sd 0:0:0:0: [sda] Mode Sense: 43 00 00 00 [861491.347535] sd 0:0:0:0: [sda] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA [861491.373854] sda: sda1 [861491.373919] sd 0:0:0:0: [sda] Attached SCSI removable disk After the physical unplug/replug, the SanDisk appeared normally again on the Proxmox host. Its Vendor/Product ID matched the existing VM passthrough configuration: USB Device (usb0): host=0781:5567 I started the VM again and Unraid is currently booting. I expect an automatic parity check because the shutdown was not completed normally. (Which is currently running) I have a few questions. 1. Does this evidence indicate that the USB boot flash device itself dropped offline or stopped responding at the Proxmox host level while Unraid was running? 2. Is the /etc/rc.d/rc.6 Input/output error consistent with the boot flash becoming unavailable during runtime? 3. Could the boot flash disappearing explain the WebGUI login failure even though the root console was still accessible? 4. Is there any known problem with Unraid 7.3.2 running under Proxmox VE 9.2.3 using QEMU USB passthrough by Vendor/Product ID? 5. Is this more likely to be: - a failing SanDisk flash drive; - a USB port/contact problem; - an xHCI controller issue; - a QEMU/libusb passthrough issue; - or an Unraid issue? 6. Would mapping the device using its physical USB port, for example host=3-2, provide any advantage over host=0781:5567? 7. Should I replace the USB boot device immediately, or first test the filesystem and move it to another USB 2.0 port? 8. Are there any specific Proxmox host logs, Unraid diagnostics or flash files that would help determine why the device dropped offline? I will attach the complete Unraid diagnostics ZIP generated after this boot. tower-diagnostics-20260715-0341.zip

Replaced my PSU for a better one. Rebuilt my drive, everything seems to be fine now. Probably my bad PSU must have caused issues in the first place. Ran an extended smart test and rebuilt the drive, 0 errors, 0 warnings. 100% success. Thanks for your help, I'm glad my unraid server is now back to normal again. I had some issues with case insensitive share names, but I found your solution in another thread. Moved the cfg file to tmp folder and re-applied settings from the GUI. Cfg was re-created and everything works now. I think this is the best software purchase I ever did, the support here is outstanding. I couldn't thank everybody enough.

Funny it says 1b.0 and I passedtrhu 1f.0. Does that mean anything? I'm not too familiarized with this subject. So I read about rebuilding the drive onto itself. That's what I need to do now to fix the array right? If so, I will first buy a new pair of sata cables before doing this process again, to avoid any new errors. Is a PSU problem even a thing to consider here? Not enough power? I have a 500w PSU (not any certified fancy font though) I'm also running an extended smart test. Short test completed with no errors. https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/replacing-disks-in-array/#re-enabling-a-disabled-disk-rebuilding-onto-itself

It was already set as AHCI. Maybe something wrong in proxmox config?

I thought I had passtrhu the devices correctly? What should I do? Is it safe do reboot unraid and try to start the array again? 02:00.0 and 04:00.0 are my NVMES used in cache. 1f.2 is my entire sata controller from motherboard

IT was solved. Probably related to that github rate limit. Everything is working fine now.

I was running a parity check and I got some read errors. After a while, my disk went offline. Parity is almost finishing but I'm not sure what I should do next to keep my files safe. Has my HD gone bad? What steps should I take to make my array protected again? I'm attaching diagnostics and some screenshots. I'm running Unraid on Proxmox. magonhacloud-diagnostics-20251104-0741.zip

it's completely random, every time I refresh they change. Something weird is going on..

Same behaviour I suppose?

I'm wondering if this issue has something to do with I'm being unable to update plugins, or even check if they are updated at all. Seems like South America is constantly getting HTTP 429 when trying to access some github endpoints. GitHubReceiving the 429 error eventhough not using github for a...Select Topic Area Question Body Hi folks; I am ot a developer but I am a just normal user of a PVE machine and the famous "PVE Helper Scripts" to create LXC containers. I am not a heavy user and on...

o meu servidor nao tem tanto uso assim. Comprei um senno vt, chegou hoje. Mas vou devolver. Barulho insuportável. Da pra ouvir de várias salas de distância, tem nem como. Foda pq eu tava hypado esperando chegar, nunca broxei tanto na vida... alguem indica um que nao faça barulho? nem precisa ser dupla conversão e top de linha, pelo jeito esses devem ser os mais barulhentos. A sala aqui é toda aberta, não tem porta. Se fizer muito barulho fodeu

It still happens. No apparent reason. Using NVIDIA GTX 1660. It works fine for several days then it just stops out of the blue. Vendor command returned unparseable data.

Well, everything worked in the past. My switch is gigabit, my desktop is gigabit and unraid says it's interface is also gigabit. The cable is new. My desktop: Unraid: About the iperf I'm still trying to figure out how to use it. Well, that was weird. MY desktop was indeed only reaching 100mbs speeds in speedtest even though windows said it was in gigabit connection. So I rebooted every router and switch and my desktop got normal speeds in speed test. Also speed transfer is back to normal. Weird. Now I dunno which answer to mark as solution

So I'm moving some files into my server and speed doesnt go beyond that. I've never had any issues regarding speed in my array. I must say I've been updating my server to latest versions but it's been a while since I used my server so I can't really tell when this started to happen. magonhacloud-diagnostics-20250308-0144.zip

Bom eu já abri um tópico aqui uma vez sobre o mesmo assunto e no final me recomendaram o Senno VT 1000va. Lendo algumas reviews falaram que tem bastante ruído porque a ventoinha fica ligada todo tempo. Alguém que tem pode confirmar? Essa parte do barulho quebra demais aqui. Eu queria um nobreak só pra poder deixar meu servidor ligado sem preocupação. As vezes a energia da uma piscada e volta e isso já dispara o verificador de paridade e demora muito tempo pra completar. Imagino que força muito o disco. Eu não consigo mais deixar o servidor ligado com medo de que qualquer falta de energia cause essa verificação enorme. Existe algum nobreak na medida do possível silencioso que funcione só para evitar esse tipo de problema? Não preciso de nada de nível empresarial, só queria poder deixar meu servidor ligado sem medo de queda de energia e que tenha o mínimo de compatibilidade com o sistema de monitoramento do unraid.