Server crashes during scheduled 'Background Work' (ref#UTPT64)

Hi! What’s not quite right with Roon?

· None of the above quite fits

None of the above quite fits

· None of these quite match

Tell us what's going on

· Is it possible to slow down the "Background Work" process?

For the past two nights, "Background Work" has been crashing my server.
"Background Work" is scheduled to run between 1:00 AM and 5:00 AM.
In the morning, I find my server burning hot with its fan running at full speed.
I can't access the server remotely, either through its web interface or with Roon Remote.
I have to unplug it, wait for it to cool down, and then restart it.
Fortunately, these two crashes haven't caused any damage.
The room temperature is 24°C (75°F).
I will provide more technical details in subsequent messages.

Tell us about your home network

· I don't think it's a network problem.

Hardware configuration:

  • Intel NUC7I7BNH Barebone PC
  • Crucial CT2K4G4SFS8213 8Go Kit
  • Transcend TS64GMTS400S SSD interne SATA III

Software configuration:

  • Roon Optimized Core Kit Version 1.0 (build 257) earlyaccess
  • Roon Server Software Version 2.70 (build 1670) earlyaccess

Crash times and last log:

  • 07/09 02:16:30 [Local 07/09 04:16:30] Trace: [updatemetadata] _SpinQueue called but work already in progress
  • 07/09 23:31:34 [Local 07/10 01:31:34] Trace: [updatemetadata] _SpinQueue called but work already in progress

Looking at the timestamps of the running log files, we see a huge amount of activity during the “Background Work” period.

11/07/2026  12:23         8411078 RoonServer_log.01.txt
11/07/2026  09:23          877130 RoonServer_log.02.txt
10/07/2026  16:09         7521875 RoonServer_log.03.txt
10/07/2026  08:28          419958 RoonServer_log.04.txt
10/07/2026  01:31          638976 RoonServer_log.05.txt
10/07/2026  01:30         8425339 RoonServer_log.06.txt
10/07/2026  01:09         8031706 RoonServer_log.07.txt
09/07/2026  10:38          385977 RoonServer_log.08.txt
09/07/2026  04:16         3293184 RoonServer_log.09.txt
09/07/2026  04:11         8435255 RoonServer_log.10.txt
09/07/2026  04:00         8435554 RoonServer_log.11.txt
09/07/2026  03:47         8435948 RoonServer_log.12.txt
09/07/2026  03:33         8435789 RoonServer_log.13.txt
09/07/2026  03:19         8435801 RoonServer_log.14.txt
09/07/2026  03:04         8435281 RoonServer_log.15.txt
09/07/2026  02:50         8460536 RoonServer_log.16.txt
09/07/2026  00:35         8490677 RoonServer_log.17.txt
08/07/2026  20:37         8470714 RoonServer_log.18.txt
08/07/2026  18:26         8471432 RoonServer_log.19.txt
08/07/2026  15:13         8459944 RoonServer_log.20.txt
11/07/2026  13:34         1360839 RoonServer_log.txt

Hypothesis: My server’s CPU must be at 100% for a very (too) long period, causing it to crash.
Hence my question: would it be possible to slow down the “Background Work”?
For now, I’ve decided to shut down my server at night, but that’s certainly not a solution because I imagine this “Background Work” is necessary.

Inadequate cooling solution violating intel’s specs at play? Properly built and working computers should be capable of running at nominal speed (100% load) without the thermal protection halting the CPU. Is your fan potentially defect or clogged? I believe it is also thinkable that the reason for the sever being unresponsive lies somewhere else and has nothing to do with the temps. Hard to tell from afar.

I had this happen too. I think it may have been due to having both background work and backup scheduled at the same time overnight. I’ve since taken background off a schedule and I haven’t had an overheating/crash since. I also now have some very quiet USB mini fans pointed at my fan-less case, just in case, but I doubt those would cool it as much as it was hot - it was hot enough to cook on! Mine is a NUC 7i5.

Thank you.
While waiting for a reply from RoonLabs, here is what I have done:

  • The backup is scheduled for 6:00 PM.
  • The “background work” schedule is now disabled.
  • I shut down the Roon server at night and when I am not at home.
    This means that, for the time being, I can no longer use Roon ARC.
    I hope this is only temporary, but I cannot take the risk of a fire.

Hi @Dirk-Pitt,

Thank you for reaching out and providing such detailed system information and log timestamps. It is incredibly helpful. Turning off the server to avoid the risk of fire was absolutely the right call!

To answer your primary question: Yes, you can slow down the background work. The background tasks typically consist of metadata updates and Background Audio Analysis. You can throttle the analysis portion by navigating to Settings > Library in Roon. Look for Background Audio Analysis Speed and set it to Throttled. This limits the number of CPU cores Roon uses, which significantly reduces the processing load.

However, we need to address the hardware side of this issue. As @BlackJack correctly pointed out, a properly functioning computer should be able to run at 100% CPU utilization without halting. While it will get warm, the CPU should automatically manage its clock speeds (thermal throttling) to prevent crashing or critical overheating.

Because you are using an Intel NUC7I7BNH, which is a 7th-generation processor, the physical cooling system has likely degraded over the years. Specifically:

  • Thermal Paste: The thermal compound between the CPU and the heatsink has almost certainly dried out, losing its ability to transfer heat.
  • Dust Accumulation: Dust can easily clog the fan and internal heatsink fins, preventing hot air from escaping the chassis.
When Roon initiates its intensive overnight background work between 1:00 AM and 5:00 AM, the CPU ramps up. Because the degraded thermal paste cannot transfer the heat to the heatsink effectively, the fan spins at maximum speed trying to compensate, but the CPU ultimately overheats and locks up the machine to protect itself.

Recommendations:

  1. Hardware Maintenance: We highly recommend opening the NUC case to carefully blow out any dust with compressed air. More importantly, removing the heatsink, cleaning off the old thermal paste, and applying a fresh layer of high-quality thermal compound is the true fix for this issue.
  2. Software Band-Aid: In the meantime, you can leave your scheduled Background Work turned off, or keep it active but ensure the Audio Analysis Speed is set to "Throttled" as mentioned above.
Let us know if you decide to tackle the thermal paste replacement, or if adjusting the settings to "Throttled" keeps the system stable enough to leave it running overnight!

Hello @vadim, thanks for your reply.
Here are my current settings regarding background processes.

I’m going to open up the NUC case to remove the dust.
However, I don’t think I’m capable of replacing the thermal paste.
I’ll see if I can find a technician nearby to do it.

You are not alone here @Dirk-Pitt

I don’t personally think hardware cooling is the issue here.

Having seen myself gather a full 20 roonserver logs in the space of two days relating to full library rescans due to dirty metadata :rofl:

I believe @DDPS has seen similar

I know @Michael_Harris had a crash the other day

Poor cooling nor lack of it is the root cause.

Seeing Roon support post about CPU maintenance prior to inspection of the logs puzzles me.

Maybe because the OP claims over temperature as the cause?

So it seems logical to suggest measures to check / restore proper CPU cooling too (first).

Yep, 100% the same.

Yeah, I see that :+1:

I saw similar during what I witnessed (which was in realtime) and my temp climbed higher than normal. I restarted Roonserver to stop what was occurring so it could have gotten worse.

Yes I fed back to another thread, as Roon Appliance on my Mac Mini hard crashed, and I had to kill the process. First time in a long time that has happened to me

@BlackJack

I agree, but we also need to find the cause:

  • Did the temperature cause the server to crash?
  • Did the server crash cause the temperature to rise?

In any case, I’ve added a temporary workaround.

Thanks @MusicD, @BlackJack, @DDPS, @Michael_Harris for your message.

I’ve just discovered this thread.
I’m going to study it carefully.

Hello @Dirk-Pitt

Hi Dirk-Pitt, and thanks MusicD, BlackJack, DDPS, and Michael_Harris for the additional data points here.

Roon 2.71 (build 1674) is live now on the Early Access branch and includes memory and performance fixes in this exact area. Could you please give it a try and see if it holds up overnight with Background Work re-enabled? Here’s more info: EarlyAccess: Roon 2.71 Build 1674 and ARC 1.81 Build 422 are Live!

If anyone else in this thread has hit the same overnight crash, please feel free to try the EA build as well and let us know how it goes either way. That’ll help us confirm whether this is the same root cause across the board.

Hello @vadim and thank you for your message.
I’m going to wait a bit before doing this test.
Indeed, I already have several problems with the version of Roon 2.71 (build 1674) on the “Early Access” branch.

Hello @Dirk-Pitt ,

Thanks for letting us know. We’ll be standing by to hear your findings when you are ready to re-enable the background work.

Hello @noris,
Thank you for your message.
Apologies for the late reply; I was away from home for a few days.
Currently, I shut down my Roon server every evening.
I plan to run a test of background work in a few days.
I will get back to you as soon as possible.

Hello @Dirk-Pitt,

Just wanted to touch base on this. We’re still standing by to hear what you find once you re-enable the background work, and whether anything changes after that. If you’ve had a chance to test it, let us know what you’re seeing, thanks.

Hello @noris, thank you for your message.
I will run the test tonight and let you know the result tomorrow (UTC+2).