Is my server under powered for docker or is my cache disk struggling to keep up?

[smakovits]

I am trying to track the source of my issue.  I am running the emby server container and it causes the web interface to eventually fail.  When emby start, I can get to the gui just fine, but after a few minutes, it begins to fail.

 

I am running the favorite, Sempron 140 and 2GB memory.  I used to run emby in docker on an intel core 2 duo 6400 and 4GB memory where things used to run just fine.  I do not use transcoding, so I only need the ability to index the media for kodi to pull in.

 

I watch the system resources watching top and htop, but the system never maxes out.  The CPU especially is typically under 30%, mayber 5-15% average.  Memory is a little different, I am using about 1.5 GB with emby (mono process) using 25% (500MB), but still, I see 300-400MB free memory.  Now I know the memory is high, but high is not max and therefore, I would not expect things to be failing the way they are.  As a result, I am hesitant to jump into an upgrade, which is why I am here.

 

The next item I thought could  be a factor is the disk where my docker img file sits.  I know the documents say use an SSD for best performance, but a spinning disk is not horrible.  I am using a raptor drive so it is not WD green disk.  This dsik is also my cache disk and 9 times out of 10 that I am need the server, the cache disk is cranking.  I can here the disk activity (read/write) even at times I don't think anything is going on, but I guess this could be because of the img file and the containers running there.  I have no idea, but I would think the disk is fine and I guess I would not expect it to be the reason the web gui stops responding, but then again, who knows.

 

Regardless, I am not sure if anything here tells us the source of the issue, I may have some memory laying around that I can toss in to see if it is a matter of running out of memory, but beyond that, it is replacing the MB and processor or the cache disk.  My old standalone ubuntu server with docker was on a spinning disk too, so it is hard to imagine the issue is the disk, but then again, the setup on unraid is a little different so maybe it is possible.  I thought I would ask first and then go get all crazy. 

 

If I was to have a plan it would be to increase memory first, then cache disk and then MB and processor, but then again, maybe it is none of these and instead I am doing something wrong.

[jonp]

I would wager this is a memory issue. If you have some laying around like you say, give it a shot and see if the issue occurs again.

[smakovits]

Sounds good, Ill take a look at what I have tonight.

[smakovits]

I would wager this is a memory issue. If you have some laying around like you say, give it a shot and see if the issue occurs again.

 

OK, great, glad it was as simple as that.  Upgraded from 2GB to 4GB memory and now things appear stable.  Ran all night and the docker containers such as Emby are still accessible.