I have been struggling with this one. Randomly, it could be hours or days my unRAID machine becomes unresponsive. The web UI is unresponsive along with the console or login via ssh. The only way to recover is to manually power off the server. For the life of me I cant tell what is causing it. I tried a Memtest and ran about 1.5 passes with no issues reported. I tried cleaning the hard drive plugs with no luck. I have had drive fails that required rebuilds, but issue still keeps happening.

This morning, I had to shut off after a crash, took my thumb drive to Windows and copied the last Diagnostic file I created late last night, along with a file I found called syslog-previous created at 2:00 AM. I see some errors at the end of this file, but I cant really understand. The entries are dated at 11:00 PM, so not sure if this is from the overnight crash, or from my last manual re-boot.

Please help. See files.syslog-previous

Thank you!

tower-diagnostics-20250810-2219.zip

Aug 9 23:59:41 Tower kernel: WARNING: Pool 'cache' has encountered an uncorrectable I/O failure and has been suspended.

Pool got suspended due to a device likely dropping offline.

Aug 10 00:00:22 Tower emhttpd: device /dev/nvme0n1 has size zero

Thank you @JorgeB This is my cache drive with appdata and all. Yesterday I did a scrub and it showed clean. It's an NVME gum stick drive.... I would hate to have to rebuild this....Any recommendations on what could be causing this? About a week ago, when booting, the system got hung up after a ZFS v.5 message scrolled by.

Power cycle the server, not just rebooting, and see if it comes back; then post new diags after array start.