Out of memory errors detected on your server

[Exes]

I am using the fix common problems plugin and I keep getting the "Out of memory errors detected on your server" error. I just bought an additional 32 gb of RAM so I'm not sure why I'm getting this.

 

Attached are diagnostics

cd1git-unraid-diagnostics-20220910-2004.zip

[trurl]

Something seriously wrong about your docker setup. 20G docker.img is usually more than enough, but you are already using 201/325G.

 

The usual reason for filling docker.img is an application writing to a path that isn't mapped.

 

Some other things about the way you have your shares setup but we can look at that later.

 

Go to Docker page, click the Container Size button at bottom, post a screenshot.

[trurl]

looks like something named 'node' is getting OOM

[Exes]

I couldn't scroll down and get a full screenshot so here is an excel file. The first ones are a little suspect

docker.xlsx

 

edit: look at the second one I posted

docker2.xlsx

Edited September 11, 2022 by Exes

[Exes]

53 minutes ago, trurl said:

looks like something named 'node' is getting OOM

I used to have a docker container named node awhile ago but I stopped messing with it after I could never get it to work. Can you see the last time it had an error?

 

I will also say the web ui crashes a lot when I click on the Docker tab due to "out of memory issues"

 

Edited September 11, 2022 by Exes

[trurl]

Do you really use all of those 300 containers?

[Exes]

15 hours ago, trurl said:

Do you really use all of those 300 containers?

Not really, I started this as a way to learn how docker containers work. I managed to clear a lot of it though.

Edited September 11, 2022 by Exes

[trurl]

https://wiki.unraid.net/Manual/Docker_Management#Re-Create_the_Docker_image_file

[trurl]

I recommend setting docker img size to 20G

[Exes]

I think I might have figured out why I was getting those errors. It looks like a few of the "gold fingers" on my HBA card are scratched off. I read on one of the threads on here from years ago that you could put electrical tape on them to fix it. Is it worth it? Or better to get a new one?

 

 

Also, would you recommend using the Docker folder instead of the img if I have a lot of containers?

Edited September 12, 2022 by Exes

[trurl]

16 hours ago, Exes said:

a few of the "gold fingers" on my HBA card are scratched off

Do you mean the contacts on the edge connector?

 

16 hours ago, Exes said:

put electrical tape on them to fix it

How could electrical tape fix a missing connection? All it could do is prevent a connection.

[Exes]

5 minutes ago, trurl said:

Do you mean the contacts on the edge connector?

 

How could electrical tape fix a missing connection? All it could do is prevent a connection.

From this thread

 

See image below. I'm not sure if it's fucked or what

 

[JonathanM]

22 hours ago, Exes said:

It looks like a few of the "gold fingers" on my HBA card are scratched off.

 

6 hours ago, Exes said:

I must be blind, I can't see any damage in that picture. All I see are the hotplug data lines being shorter so they make contact only after the card has power. Maybe a clearer closeup with the damage circled?

 

The referenced thread is about complete boot failure, so if your system is booting to Unraid that's not the issue.

[Exes]

I'm just wondering if the HBA card has anything to do with the out of memory issues. If not, then I'll close the thread