Hi all,

I built a new Unraid server and have been running into stability issues since setting it up. The server becomes unresponsive at random intervals.

Sometimes it will run for 14 hours, other times it becomes unresponsive within 15 minutes. When it freezes, I lose access to the web UI and all running services (Radarr, Sonarr, etc.) and the only way to recover is by forcing the server to restart by holding down the power button.

I have run Memtest86 for 24 hours (18 passes) with no errors detected and all temps seem fine when the system is running.

System Info:

• Unraid version: 7.1.4

• Intel i5-14600K

• ASRock B760M Pro-A WiFi

• Corsair Vengeance DDR5 64GB (5200 MHz)

• Corsair RM850x (80 Plus Gold, ATX)

• Cache drives: 2X Samsung 970 EVO Plus 1TB NVMe

• Storage: 3X WD Green HDDs (1TB + 2X3TB)

I am new to Unraid, so I am not sure what further troubleshooting steps I should be taking to narrow down the cause. Is there something that I can do to test whether it could be CPU or motherboard related?

I have attached diagnostics taken after the latest reboot.

Thanks in advance for your help!

tower-diagnostics-20250927-1415.zip

Edited September 28, 2025Sep 28 by Buzzman

Setup Syslog Server.

Thanks for the quick response. I have setup Syslog Server with the following settings and will post updated logs from the next time that the server freezes.

Will the Diagnostics will retrieve these logs?

Edited September 28, 2025Sep 28 by Buzzman

2 hours ago, Buzzman said:

Will the Diagnostics will retrieve these logs?

Nope, you need to get it from the syslog share.

45 minutes ago, JorgeB said:

Nope, you need to get it from the syslog share.

It should also be in the diagnostics as well since the Mirror to Flash option is set as well as putting the serves address into the Remote Server field. Normally you only set one of those two fields.

Thanks for the replies, I really appreciate them.

19 hours ago, itimpi said:

It should also be in the diagnostics as well since the Mirror to Flash option is set as well as putting the serves address into the Remote Server field. Normally you only set one of those two fields.

OK, thanks for the advice. I will update my settings following troubleshooting.

My unraid server has crashed several times since, with the longest it has been running approximately 30 minutes. I am struggling to even start the array to access the syslog folder. I managed to get a couple of sets of diagnostics and a copy of the syslog.

There seems to be CPU-tainted errors but I am not sure how to determine the cause of these errors. Are there any ways to check the most likely cause of these errors?

syslog-192.168.4.48.logtower-diagnostics-20250929-1433.ziptower-diagnostics-20250929-1458.zip

2 hours ago, Buzzman said:

with the longest it has been running approximately 30 minutes.

That strongly suggests a hardware issue, since you have multiple RAM sticks, try using the server with just one, if the same try with the other one, that will basically rule out bad RAM, and if it still crashes, the CPU would be my next suspect. Since it's a 14600K, could be affected by the Intel 13/14 gen issues.

23 hours ago, JorgeB said:

That strongly suggests a hardware issue, since you have multiple RAM sticks, try using the server with just one, if the same try with the other one, that will basically rule out bad RAM, and if it still crashes

OK, thanks for the suggestion. I tried running the server with just one stick of RAM and it become unresponsive both times while testing each stick of RAM.

I have attached the latest set of diagnostics and copy of the syslog which seem to contain the same types of messages.

23 hours ago, JorgeB said:

the CPU would be my next suspect. Since it's a 14600K, could be affected by the Intel 13/14 gen issues.

Is there a suggested method to test the CPU and confirm whether it is CPU or motherboard related?

I did a two hour stress-ng test and didn't see any errors using the below command.

docker run --rm --name=stressng --cpus=0 --memory=2g ubuntu:22.04 \

bash -c "apt-get update && apt-get install -y stress-ng && \

stress-ng --cpu 0 --cpu-method fft --timeout 10m --metrics-brief"

Edited September 30, 2025Sep 30 by Buzzman

2 hours ago, Buzzman said:

Is there a suggested method to test the CPU and confirm whether it is CPU or motherboard related?

I believe there's an Intel testing tool for Windows, but not sure how reliable it is

Thanks for all of the help and suggestions. I really appreciate it.

I looked into the Windows Intel testing tool and the feedback seems to be that it can miss issues and give false positives.

Is there any other testing that I should bother trying such as running in Safe Mode? Or would that likely not reveal anything based on the errors that I have been seeing in the logs?

If it is likely to be a CPU issue and I need a replacement, are there any suggested Intel or AMD CPUs that I should look at? I hadn't heard of the 13/14 gen issues previously but it sounds like I should completely avoid these.

You can try safe mode to rule out any plugin issue, but likely that's not the issue here.