Hello, this trial period of Unraid has been tumultuous at best. After multiple reset and disk wipes and coming to the same issue, I need some help.

 

I've been wanting to stress test the server for 3 days before attempting to install any apps again since I thought that was my problem last time, however this time I have not done anything to the server, no dockers, no vms, no data, and it won't last for 24 hours before having problems.

 

When it starts having issues, I can only login via the assigned IP address, tower.local does not work. This behavior is replicated in other browsers, regardless of if I'm connected to local wifi, ethernet or using a vpn to connect via cellular. In addition to only being able to connect via the assigned IP address, my array list/details and dashboard are completely blank. When I shut it down or reboot it in this state, it does so uncleanly, which triggers a parity check on the next boot. Another weird detail is that when I initiate a powerdown via CLI (since I can't in the webgui), the array details will briefly appear on the webGUI like normal as it attempts (and fails) to shut down cleanly, dashboard remains blank though.

 

I can see and access the shares/settings etc. and running htop via CLI shows normal CPU and RAM activity so it seems like everything is working, but I'm sill fairly green when it comes to CLI usage so dunno if there's more I need to check. I ran a dedicated MEM test and the RAM passed. rebooting seems to fix the issue temporarily, but again, it's unclean and annoying starts a parity check and fails within 24 hours.

 

I've tried:

-wiping the USB and re-imaging w/ unraid OS

-wiping all drives separately/externally and then above step again

-Made sure I've used USB 2.0 port

-Reset and updated motherboard BIOS

-Reseated, cleaned and repasted CPU (again), verified all CPU pins on MOBO were good.

-reseated and tested RAM 

 

Attached are my diagnostics file, I reviewed the logs and saw some "processor segfaults" which, I'm not sure if that is indicative of of a failing CPU or not, but the CPU was running great in its previous life as a gaming CPU, so I dunno. (If anyone has a recommendation on a dedicated CPU tester I can run to help rule that out, it'd be appreciated.) I really wanna like unraid, but this has been a nightmare.

 

 

tower-diagnostics-20241106-0729.zip

There are multiple apps segfaulting, since memtest is only definitive if it finds errors, and you have multiple sticks, try using the server with just one, if the same try with a different one, that will basically rule out bad RAM.

On 11/6/2024 at 8:40 AM, JorgeB said:

There are multiple apps segfaulting, since memtest is only definitive if it finds errors, and you have multiple sticks, try using the server with just one, if the same try with a different one, that will basically rule out bad RAM.

replaced/upgraded both sticks rather than chasing down a bad one, the cooler is bulky and covers the ram so re-doing that means repasting the CPU each time. boo.

 

however since putting the new sticks in I have no segfault errors anymore, which typically were visible in the diag log after boot, gonna mark as solved while I monitor  for a few days and will troubleshoot the mainboard/cpu if it continues. Thanks!