I have a single node k3s instance that s kinda freaking out rancher-users #k3s

Title

freezing-lunch-86527

01/11/2023, 5:22 AM

I have a single node k3s instance that's kinda freaking out 😄

k3s server

is taking like 200% CPU and the kubernetes API is becoming unreachable which is causing a bunch of stuff to crash which is making the issue worse. I've poured through

journalctl

and don't really see anything notable. Any ideas how to track down the root of this?

creamy-pencil-82913

01/11/2023, 6:22 AM

Maybe look at what pods you have running that would be making a ton of requests to the apiserver? There are no significant errors or warnings?

freezing-lunch-86527

01/11/2023, 6:24 AM

lots of errors timing out hitting the API server.

I think i sort of figured it out. Seems like the instance was resource starved. Strange because I didn't add anything new recently... but I think once things started failing a bunch of stuff was crashing and using more than normal resources upon startup which was causing other things to crash and so on

creamy-pencil-82913

01/11/2023, 7:01 AM

if you are running with swap and don’t have any memory limits on pods, you can definitely push things to the point where the performance becomes pathological

best practice is to turn off swap, but k3s lets you use it because it can be necessary on some smaller systems

freezing-lunch-86527

01/11/2023, 3:17 PM

🤔 honestly don't know if I have swap setup. I'll check on that. Thanks for the tip

#k3s Join Slack