Out of memory errors detected on your server

[be4con]

Hi all

 

I've been suffering occasional Out of Memory Errors for a while and haven't been able to get to the bottom of them myself. Can anyone help me resolve them?

 

I haven't noted any issues with this until recently, then the last few days my MyMedia for Alexa Docker keeps stopping and won't auto-restart so I'm wondering if that is related.

 

My diagnostics are attached.

 

TIA

 

tower-diagnostics-20230619-1126.zip

Edited June 19, 2023 by be4con

[JorgeB]

If it's a one time thing you can ignore, if it keeps happening try limiting more the RAM for VMs and/or docker containers, the problem is usually not just about not enough RAM but more about fragmented RAM, alternatively a small swap file on disk might help, you can use the swapfile plugin:

 

https://forums.unraid.net/topic/109342-plugin-swapfile-for-691/

[be4con]

Thanks, it happens periodically, maybe once per month on average. How do I go about limiting the RAM for VMs and/or docker containers?

 

Do you think this could be the reason that the MyMedia docker keeps stopping, or should I look for specific help with that docker community? 

[JorgeB]

https://forums.unraid.net/topic/57181-docker-faq/?do=findComment&comment=566088

 

[be4con]

2 hours ago, JorgeB said:

https://forums.unraid.net/topic/57181-docker-faq/?do=findComment&comment=566088

 

Thanks, I have only 4G RAM. Is it reasonable to limit it to 1G max per Docker? 2G??

[JorgeB]

4GB is the minimum for NAS only, you may just need more RAM (or use swap plugin).

[be4con]

From what I had read, I thought 4Gb should be plenty, but I've managed to find an upgrade kit for my ancient server, so I'll boost the memory to the 16G max and see if that resolves everything. Thanks.

[be4con]

Still having issues. 

 

I upgraded my RAM to 16Gb, and I also limited the memory use to 4Gb per docker as advised. I have yet to institute a swap file, but before I do that I thought we should explore further why this might be happening. I'm not running anything especially RAM heavy - I only have 2 active dockers, Delugevpn and MyMedia for Alexa - and there should be plenty of available RAM for what I am trying to achieve without adding a swap file surely?

 

Any further advice before I explore swap files? 

[itimpi]

I believe tht the issue is often memory fragmentation rather than total memory, and I think the swap file option helps with this.

[be4con]

I've added the swapfile plugin and started it with 8MB file.

 

The error has gone for the moment but as soon as I did this I got 2 new errors reported, not sure if correlation or causation:

 

[itimpi]

3 hours ago, be4con said:

I've added the swapfile plugin and started it with 8MB file.

 

The error has gone for the moment but as soon as I did this I got 2 new errors reported, not sure if correlation or causation:

 

The second one is the worrying one - it implies you have something writing to a location that is RAM.

[be4con]

Both errors disappeared when I uninstalled the swap file plugin, but the Out of Memory error recurred. Any suggestions on what I do?

[JorgeB]

Did you reboot? Might be seeing an old warning, or post new diags.

[be4con]

Yeah I rebooted both after the install of the swapfile plugin and after uninstall of it, so the warnings were new at each stage. Just run FCP again and got a new error:

 

Fresh diags attached
 

tower-diagnostics-20230704-1808.zip

 

 

Edited July 4, 2023 by be4con

[JorgeB]

Diags are not available, please try again.

[be4con]

tower-diagnostics-20230704-1808.zip

[JorgeB]

Memory cgroup out of memory: Killed process 20657 (MyMediaForAlexa)

 

Yeah, still happening, may be a badly behaving container, or one with a memory leak, if you stop that container do the OOMs stop?

[be4con]

Thanks

 

Oh, it's definitely misbehaving. I've got another support chat running about that container, although the support thread for the docker is pretty much dead. I've been having problems with it constantly stopping and refusing to autostart. So, I can't really check whether stopping that resolves anything because more often than not it's in the stopped state.

 

I went through the process of uninstalling that container, deleting the image and the container's appdata folder and performing a fresh install but the issue with it continually stopping just returned as soon as I reinstalled it, and it's not really behaving normally since I reinstalled it either (it won't index the files on the server). I raised a ticket with the MMfA devs and they have failed to respond to that too. Getting a bit frustrating to be honest.

[be4con]

Any suggestions please? This is driving me nuts