I started getting seg faults on my unraid server. I have tried replacing the ram but the faults stay the same. Another oddity, I am not able to shutdown the server. If I use the button on the GUI 90% of the time nothing happens and if i use the CLI it starts but eventually starts getting call traces and never shuts down.

leviathan-diagnostics-20250928-1250.zip

Edited September 29, 2025Sep 29 by chishola1

2 hours ago, chishola1 said:

I started getting seg faults on my unraid server. I have tried replacing the ram but the faults stay the same. Any idea on how to solve this?

leviathan-diagnostics-20250928-1250.zip

If it's only segfaulting with the python script you're running, start there

What is the best way to find/disable that python script?

5 minutes ago, chishola1 said:

What is the best way to find/disable that python script?

Figure out which container is running a python3 instance. running ps -auxf can help you narrow it down

8 minutes ago, Michael_P said:

Figure out which container is running a python3 instance. running ps -auxf can help you narrow it down

Could you help me understand the output of that command?

I just checked my log again,

Sep 29 08:31:16 Leviathan kernel: python3[28091]: segfault at 0 ip 0000000000000000 sp 00007ffc06550728 error 14 likely on CPU 6 (core 12, socket 0)

Sep 29 08:31:16 Leviathan kernel: Code: Unable to access opcode bytes at 0xffffffffffffffd6.

Sep 29 08:32:00 Leviathan kernel: .NET ThreadPool[30288]: segfault at 9 ip 0000153a4357eb88 sp 0000153a18eb50d0 error 4 in libclrjit.so[17eb88,153a43400000+322000] likely on CPU 6 (core 12, socket 0)

Sep 29 08:32:00 Leviathan kernel: Code: a0 45 31 ff 41 0f b6 c1 89 45 c4 44 0f b6 6d ac 4d 85 f6 75 2d 48 8b 45 a0 4c 39 e0 44 8b 75 90 0f 84 a3 03 00 00 41 83 c7 40 <4d> 8b 74 24 08 49 83 c4 08 4d 85 f6 74 e1 66 2e 0f 1f 84 00 00 00

Sep 29 08:33:02 Leviathan kernel: python3[38176]: segfault at 0 ip 0000000000000000 sp 00007fffe0625b88 error 14 likely on CPU 6 (core 12, socket 0)

Sep 29 08:33:02 Leviathan kernel: Code: Unable to access opcode bytes at 0xffffffffffffffd6.

Sep 29 08:33:17 Leviathan kernel: deluge-console[39808]: segfault at 0 ip 0000000000000000 sp 00007fff162e8e68 error 14 likely on CPU 4 (core 8, socket 0)

Sep 29 08:33:17 Leviathan kernel: Code: Unable to access opcode bytes at 0xffffffffffffffd6.

Sep 29 08:33:17 Leviathan kernel: deluge-console[39849]: segfault at 0 ip 000014e542d799ed sp 00007ffd154f1bb0 error 4 in libpython3.13.so.1.0[1799ed,14e542c3a000+2cc000] likely on CPU 6 (core 12, socket 0)

Sep 29 08:33:17 Leviathan kernel: Code: ff ff 0f b6 dc 41 89 c2 0f b6 c0 4c 63 fb ff 24 c6 f3 0f 1e fa 48 8b 85 b8 fe ff ff 49 8b 7c 24 f0 4d 8b 7c 24 f8 4c 89 68 38 <48> 8b 07 49 83 c5 02 85 c0 78 27 48 83 e8 01 48 89 07 75 1e 48 8b

Sep 29 08:34:07 Leviathan kernel: python3[44547]: segfault at 0 ip 0000000000514e39 sp 00007ffff7c8a5c0 error 6 in python3.9[114e39,41f000+289000] likely on CPU 6 (core 12, socket 0)

Sep 29 08:34:07 Leviathan kernel: Code: 48 85 ff 0f 84 12 6a 00 00 48 8b 47 38 48 85 c0 0f 84 05 6a 00 00 4c 89 fe 44 89 4c 24 40 4c 89 f7 ff d0 44 8b 44 24 40 89 c6 <49> 83 2f 01 0f 84 be 4d 00 00 49 83 2e 01 0f 84 f8 4c 00 00 85 f6

looks like it is not just python

CPU could be faulty, have you removed it lately?

Try turning off your deluge container, too - it looks like that's what's running the python process

I removed it yesterday, but that was while troubleshooting this issue. The server had been up for around 30 days and then just died. No changes were made during that time, hardware or software.

I will throw a windows hdd in and run the intel Processor Diagnostic Tool

3 minutes ago, chishola1 said:

I removed it yesterday, but that was while troubleshooting this issue. The server had been up for around 30 days and then just died. No changes were made during that time, hardware or software.

Check to make sure you didn't bend any pins on the the LGA

It was doing this before I pulled it but I will verify that there are no damaged pins.

I see that you have a 13900K, that's one of the most affected models by the Intel 13/14 gen issue, a BIOS update can sometimes help, if the CPU is not too far gone.

Already updated to current version

If the diagnostic tool shows that the processor is bad what would you recommend replacing it with without having to replace MB?

Typically the problem is just the CPU.

6 minutes ago, JorgeB said:

Typically the problem is just the CPU.

I'm asking for a recommendation for a 13/14 gen cpu if I find that my 13900k is bad.

You can get another one of the same model, if that is what you need, if a less powerful CPU is enough, get a non K model; typically there are no issues with those.

After not being able to get windows to load I replaced the CPU and it seems to have fixed it. Time will tell but I am not seeing any errors in the log and I am able to shutdown the machine properly again. Thank you for the help.