What’s happening?
· Other
How can we help?
· None of the above
Other options
· Other
Describe the issue
Excessive memory use in last update
Describe your network setup
QNAP TS-873A, 1GB Ethernet, Orbi 6 wifi
Is anyone having memory bloat with the most recent update on Linux? I’m seeing the Roon Server now taking 50% of my NAS memory. After restarting the Roon Server, the memory usage dropped to about 1.4 GB.
Roon Server version 2.51 build 1534, running on TS-873A QTS 5.1.9, RoonOnNAS install.
-Alan
Hi @Alan_Sparks,
Thank you for reaching out to Roon Support!
Could you please confirm whether you have any historical data on memory consumption for the NAS in question?
The goal is to determine whether the issue is directly related to a recent Roon update or if it’s part of a memory leak behavior within certain NAS implementations—a topic that has been discussed extensively in our community forums.
Any additional context you can provide will help us better isolate the root cause.
I’m seeing this too on a vanilla ubuntu server 24.04. With all services running the memory load on the machine was typically only around 3 to 3.5 GB, I just checked and it was around 6.5 GB and when i killed roon server it dropped to 2.7 GB. I’ve never seen the total memory load at more than 4 GB - I have a very small library (< 10K tracks). I generally look at memory load every few days. Restarted roon and it returned to about 3.5 GB, so it looks like you may have a memory leak.
I have 7760 tracks in the library. It is historically significant, I’ve not seen more than about 3 GB RAM used by the app until this. Didn’t realize it until memory alerts went off. I’ve restarted the app and am monitoring, it’s now 1.13 GB and not growing much… haven’t used Roon in the last few days though.
It looks like memory leaks are back (Ubuntu Server 24.04). You can see on the attached graph the memory resets from Roon updates and other reboots. Memory was stable until early May, but then you can see steady increases between reboots, steepest since the last update.
1 Like
Hello @Fernando_Pereira @Alan_Sparks @Patrick_Conroy,
It’s been a few days since we all restarted our Roon servers — has anyone noticed whether the RAM usage is increasing again?
It would be great to compare observations to determine if the issue is recurring.
It still creeps up, but maybe more slowly.
Yes, I restarted the roon server last night, immediately after restart htop was showing 3.8% memory usage, an hour or so later it was up to 4.5% and this morning it is now showing 21.4% of memory (or 5.1 GB). At no time since restarting have I played any music - I did run one of the remotes on Windows for about 2 minutes.
Have been watching on my NAS over last 4 days of uptime. Looks like the RAM usage of the process has been increasing, from about 1G to 3G. Has been mostly idle.
I have Toon running on DietPi and I have 75k tracks in my library (73.5k are local files).
RAM use has steadied out at about 4gb.
Unless folks are experiencing crashes or severe performance issues, I wouldn’t worry about things.
If a slight higher RAM use gives a better Roon experience, then I’m all for it.
I only have 8gb RAM with a N95 CPU, and my performance is acceptable even with DSP for separate channel PEQ, headroom and volume levelling.
Hello,
Thank you for the update.
Based on the screenshots provided and the information regarding the library size, we can confirm that this behavior is considered normal for the product. Roon performs several background tasks such as metadata analysis, audio fingerprinting, artwork retrieval, and library indexing, which can temporarily increase resource usage. These processes are essential for enhancing your library’s accuracy and user experience.
I think there’s “normal,” and “historical normal.” Having run Roon for a couple or years or so, I’ve not before seen the server hit 6 GB RAM usage. I’ve noticed it has grown again since my last post, doing its own thing. But honestly, this wasn’t temporary, it didn’t release the RAM till it was killed. Guess time will tell.
Certainly let us know how things look heading into next week @Alan_Sparks, thank you!
After running mostly idle for 10 days, my server process ballooned to 6.5G memory usage, after idling between 1-2 GB or so. I’ve seen it wobble back and forth, but this was a massive uptick. Still responding to clients though.
An hour later the RAM usage grew another GB and started putting some pressure on the NAS.
I downloaded the logs from the RoonOnNAS app before restarting the Roon Server. That knocked the RAM usage way down.
Let me know if you want any of those logs.
-Alan
Is anyone taking this seriously?
Since my post above, RoonServer bloated again. It’s by far the largest process running on the NAS at 9GB.
I had to restart the Roon process to free up half the NAS’s memory. This while unused, only its own threads. This seems crazy for a library of 7760 tracks. Certainly didn’t do this before about the last release.
-Alan
Hi @Alan_Sparks ,
Thanks for the report. We were able to enable diagnostics mode for your account and what this did was upload the error log to our servers for review, so no need for you to manually send it over. Looking over the error logs, we did notice the memory spikes, but it doesn’t appear to have happened in the last two days, is this accurate from your point of view as well?
When the memory issue last occurred, it seemed to happen right after a Qobuz sync, I wonder if this could be related to Qobuz. Next time you experience this issue, can you try to temporarily disable Qobuz via Roon Settings → Services and see if the issue persists?
If you look at my postings above, you might notice that it can take a few days for the growth to start, then when it does it picks up speed fast. How often does “qobuz sync” happen?
I restarted the server due to your yesterday release. It’s now taking about 1GB RAM. I continue to monitor.
-Alan
Hi Everyone,
We’ve created a ticket with our development team to address this issue. While we don’t have an exact timeline to share just yet, please know that it’s actively being worked on.
We understand how frustrating this has been and truly appreciate your patience. We’ll be sure to keep you updated as soon as we have more information.