"Your server has run out of memory, and processes (potentially required) are being killed off. You should post your diagnostics and ask for assistance on the unRaid forums"

 

This is an error I noticed when checking "Fix Common Problems" today. I was unaware of this error, or when it occurred. Everything seems to have been fine lately. But I was having major issues not long ago. Maybe unraid is still having problems, but its happening at night when I don't notice it.

 

The only change I've made to my server recently is a switch from an LSI 9211-8i with an HP Expander, to an Adaptec ASR 71605. This change seems to have been purely beneficial from what I can tell. Performance and stability seem to be great. and I no-longer have drives randomly disappear like I did with the LSI card.

serverus-diagnostics-20200901-1307.zip

Yes lots of OOM in your syslog.

 

Maybe unrelated but why do you have 50G allocated to docker.img? Have you had problems filling it? Maybe you changed whatever was filling docker.img so it is filling RAM now.

I used to have issues with it filling, but that was a while ago when I didn't properly understand docker. I haven't had that issue is a while, but I havent see any reason to shrink my docker img back. 50G isn't hurting me.

 

The only process that should be able to write into my ram is the Plex transcoder, and I changed it to "/dev/shm" instead of "/tmp" specifically so it cant take all my ram.

So is there any way for me to find out whats causing the OOM errors? During the day everything seems fine, I havent noticed anything wrong. If they are happening at night then it could be the scheduled tasks in Plex running at the same time as the mover and parity checks.

 

EDIT: I take back what I said. Im in my WebUI right now and its barely responsive. I have to refresh the "Dashboard" tab multiple times to get everything to show up. The "Main" tab will not load completely, and neither will the "Plugins" or "Docker tabs"...However all of my containers are running just fine, not even a little slowdown. This is definitely a new issue.

Edited September 2, 20205 yr by relink

Ok yah, I must have just not noticed because Dockers are running fine. But there is definitely something very wrong. I have uploaded a diagnostics from before and then after a reboot. Even after a reboot something isn't right. I stopped the array, so nothing was running, and something was still very off, the whole UI is incredibly slow. To clarify I had to reboot from the command line over SSH because the buttons in the WebUI were unresponsive.

serverus-diagnostics-20200902-1014.zip serverus-diagnostics-20200902-0941.zip

Have you tested in SAFE mode and / or without dockers/VMs?

I have not tested safe mode, but I have tried it with Dockers and VMs disabled.

Ok this is going to sound ridiculous...I think it may actually be the macOS version of Firefox.

 

I primarily access my unraid server through my laptop which is a hackintosh running macOS Catalina. Anyway, I was on my phone and I accidentally tapped on my unraid tab in my browser and poof it loaded almost instantly. I then connected to my Windows 10 desktop (with a probably very outdated version of Firefox) and tried to load my unraid server, again, it loaded almost instantly. I then went back to Firefox on macOS and it was the same crap show as earlier. I then opened Safari on macOS and logged in there and it loaded instantly...