Serviio

I saw the 2016 post about heap space problems, and now that I have 1.9.2 installed, I'd like to know if I'm going about this the right way. I'm kind of a noob, so please bear with me.

I'm using Fedora 28. I'm running Serviio as root, though there were some very old posts that said you can't do that. Seems to work for me.

Serviio started ok for me, then died of a heap attack within minutes. I tried setting environment variables to set the heap space, but that failed. Maybe because it's specified explicitly in the startup script??

I edited serviio.sh increasing the -Xmx and -Xms values. Each time it would run a little longer. I finally got medieval on it and set those parms to large values:



So now it seems to run, and stay running. I realize I could have changed the values I used to 3G and 2G instead of leaving them as mbytes.

I have 16,200 music files, 5300 video files, and 8300 images (photos.)

The music is in 1450 directories. There's 380 dirs under video. The photos are in 1400 directories, in a CIFS share on another machine mounted here.

I got those stats using a simple command, which I believe to be more-of-less accurate.

Am I doing this right?

Another question: When I launch serviio by typing it detaches from the console, as I expect. But all the error messages seem to flow to the console that launched it. Once that console session is disconnected, those messages don't seem to appear anyhere in any log file in serviio/log or /var/log.

Do I need to launch it with redirection to a different log file (for both stdout and stderr) if I want to see them?

Thanks!
-Tom

Maybe there is a dodgy image that causes the heap space fill up. Check the log if it fails at around the same time

I suppose it's possible, but my observation is that it runs longer the more memory I throw at it. Once I give it 3G and 2G respectively, it runs indefinitely.

Now that it has digested the huge number of files I originally threw at it, I reduced it to 1G and 2G. It seems to be stable that way.