Yes, confirm this as well with the latest version running on a Intel NUC (Linux).
Well, I was getting so frustrated by not being able to use Roon because of this memory leak issue that I looked whether there was a docker image available through Container Station and there is. Just installed it with setting a limit of 8GB for the docker image and till now no memory issue at all. Must say the migration went easy: just put a backup back in the image and after that change the Paths to my music collection and backup location in Roon as the paths are different in the docker. Maybe a good tip for other people having the same frustration.
I’ve also set a reboot script for 4am each morning so when I use it in the day I’m getting about 3-4 hours out of it each day before I switch to vinyl. Hopefully a fix is incoming
Yes, good point, done.
THIS.
This is what I think I am seeing. Even with no load, my RoonAppliance hovers at 30-40% of the CPU. It’s doing this.
I’ve brought it up on here before but the suggestion was that my CAs and Cast Groups were disconnecting and reconnecting. They are not. (I have logs on my in-house connections - the CAs don’t connect/disconnect/connect. They are pretty solid.
I’m guessing there is something wrong in the RoonAppliance endpoint that keeps the service doing device discovery on cast devices.
I’ve managed to get the load down on the CPU by turning off memory compression in my NAS, I’ve also noticed desktop remote app seems to hit it much harder, particularly if you leave it open…so for now I’m using it, q’ing a bunch of stuff and then closing it. I’m roughly able to get a full day out of it and the reboot at night is resetting things ready for next day.
hopefully a fix comes soon cos it’s a real pain to have to use the remote so sparingly.
Read also:
Here’s the thing about memory compression - it’s not likely to cause a problem. It’s a sound and mature technology that comes from the Linux zram kernel module. It has very little CPU performance overhead, and can have a tremendous positive impact on disk performance. Here’s the key thing: It only compresses data that isn’t being used very often! By definition, it’s not going to do much with the stuff everybody on these pages is concerned about. I see people doing this all the time with Synology devices, achieving a placebo effect for folks. That’s why it is on by default.
Ive been reading about this for a while and noticed its impacting me also. The roon process was using 10GB ram on my nas and after reboot and a day later its sub 2GB.
Absolutely BlackJack! I just unchecked my Synology 918+ memory compression about 6 hours before bumping into your post here and sure enough, it is maintaining well without the prior issues of “Waiting for Roon Core…” with a memory utilization of 90% (which is crazy). I had been having this issue for about a week now (seemingly after some Roon updates) and it was driving me nuts. To add insult to injury, I’ve got 2 SSD NVME 1GB drives and 16GB of RAM and had never had this problem before. It’s looking like that is the solution. Fingers crossed!
UPDATE: Spoke too soon. Memory consumption has been rapidly growing since this post and is not at 10.1GB (67%) and climbing again without anything playing. Unfortunately the unchecking on the memory compression didn’t do the trick as I hoped. The strange thing is that I’ve been using Roon the same way with the same equipment for almost a year now without issue and seemingly after the recent updates in May 2022, the memory problem started. The quest for a solution continues.
I have 8GB of RAM in my 918+ and don’t have any performance issues with a 2TB library. One thing to keep in mind is that the 918+ is not spec’d to work with more than 8GB of RAM, even though people have discovered that it seems to work. But that overextension might actually be part of the problems you are/were seeing.
Interesting. I never knew that since the 918+ allowed me to put in the 16GB of RAM and never really had an issue before. Then again, maybe it was never “pushed” like Roon is pushing it now. I do see that you are correct in that it is only spec’d for 8GB of RAM. I’m going to try to downgrade it to 4GB to see if I noticed a performance improvement or change as opposed to having the 16GB in there. Thanks DDPS.
24 hours later and its gone from sub 2GB to 7.5GB. Not sure if this helps anyone but figured I would report it.
On my i3 NUC I seem to be having a similar problem. RAM use just increases over time until after a number of hours, Roon becomes unresponsive and has to be killed and restarted. Found this thread looking for a solution…
I’m running Qnap NAS for the library with i3 NUC (8gb RAM) for Core.
Definitely some problem here that didn’t used to exist. My Roon core was again unresponsive (as it becomes every other day now) and I checked the task manager. It was using nearly 6gb of RAM and also 60MB/sec disk resources. System was nearly unresponsive due to such heavy load and returned to normal when I killed Roon application. Opening it again, and the whole process starts all over until I’m here again tomorrow or the next day. I’ll start a new report ticket with Roon support.
I’m having the same exact problem. After Roon Core maxes out the memory consumption and then the NAS memory collapses shutting the Roon Core down where you can see the available memory immediately become available within a couple seconds after that. And like you said, the process repeats itself a few times a day. Never happened before until the recent updates in May 2022 (for me at least). Dying to figure out what the fix is…
My core was running really slow and went to check my NAS resources. Roon was using almost 10GB of Ram.
Currently its using 10.22GB, is this normal?