Hi everyone,

I am currently struggling with a recurring and very frustrating issue on my Unraid server and would really appreciate any advice.

I am running Plex in a Docker container using plexinc/pms-docker, with Intel iGPU (Quick Sync) enabled for hardware transcoding.
Almost every day, during Plex scheduled maintenance tasks, the server suddenly reboots — it looks like an instant hardware reset, as if the reset button had been pressed.
This typically happens after 30–40 minutes of Plex running its maintenance tasks.

In earlier cases I also experienced full system freezes, but lately it has mostly turned into instant reboots.
Unfortunately, the syslog always stops about 10–15 minutes before the reboot, so no kernel panic or other obvious error is logged.

What I have already tried:

• Limited the Plex Docker container to 8 CPU threads and 8 GB RAM

• Performed Live Memory Tests (no errors found)

• Checked syslog (no OOM killer, no kernel panic, no obvious issues)

• Monitored drive temperatures and health (no issues found)

• Running Unraid 7.1.3

• Using the official plexinc/pms-docker container

This behavior seems highly correlated with heavy I/O and GPU load caused by Plex maintenance.

I would appreciate any suggestions regarding:

• Further debugging steps (kernel log settings, hardware tests, extra logging on USB, etc.)

• Known issues or limitations when using iGPU (Intel Quick Sync) with plexinc/pms-docker on Unraid

• Links to similar forum threads or external guides

I will attach my latest diagnostics ZIP to this post.

Any help or guidance would be greatly appreciated.

unraid-diagnostics-20250613-1401.zip

Since a couple of days my Plex crashes immediately after startup... tried restoring the database but that does not make a difference...

Just to be sure I reverted to the previous unraid version, did not make a difference..

I deleted plex, including the folder on appdata, and resetup the docker.. seems to work now.. its repopulating without crashing.. did not setup the quicksync link though.

Unfortunately, this issue has been present for me for a few months now. I have already tried binhex-plex, binhex-plexpass, the Linuxserver Plex version, and finally the official one. Each time, a new container and new appdata were created accordingly. I have tested it with and without QuickSync — always the same behavior. The problem has persisted across several Unraid versions.

Hard shutdowns/reboots are usually the power supply, especially if it only happens under load. Next on the list is the CPU either overheating or failing, then the motherboard.

2 hours ago, Michael_P said:

Hard shutdowns/reboots are usually the power supply, especially if it only happens under load. Next on the list is the CPU either overheating or failing, then the motherboard.

I’m quite certain this is not a hardware issue. I already had the same problem with my previous case, which had a different cooler, different motherboard, and different power supply. The only component that hasn’t changed is the CPU — and I previously used this CPU in my main PC without any issues.

Also, the CPU cores are heavily limited, and during the Plex maintenance tasks, Unraid runs at only around 60% total CPU usage.
In addition, I have other very CPU- and IO-intensive processes running on the system as well, and none of them cause any issues — the system is completely stable under those loads.

If I had to guess, I’d suspect that something related to the iGPU and FFMPEG might be causing this — but I don’t have enough knowledge in that area to say for sure.

Has anyone seen similar behavior or knows what I could check next? Are there any good logs or debug options I could enable to get more insight into what’s going on?

15 hours ago, KaNaCl said:

I’m quite certain this is not a hardware issue. I already had the same problem with my previous case, which had a different cooler, different motherboard, and different power supply. The only component that hasn’t changed is the CPU — and I previously used this CPU in my main PC without any issues.

Also, the CPU cores are heavily limited, and during the Plex maintenance tasks, Unraid runs at only around 60% total CPU usage.
In addition, I have other very CPU- and IO-intensive processes running on the system as well, and none of them cause any issues — the system is completely stable under those loads.

If I had to guess, I’d suspect that something related to the iGPU and FFMPEG might be causing this — but I don’t have enough knowledge in that area to say for sure.

Has anyone seen similar behavior or knows what I could check next? Are there any good logs or debug options I could enable to get more insight into what’s going on?

If you're willing to dismiss a hardware problem outright (and you mention it's the same CPU), then you're pissing up a rope. You'll probably should ask for help in the Plex forums to help you debug since it's not an unraid issue.

How full are your disks/pools?