3

I have a 3-node ubuntu microk8s installation and it seems to be working ok. All 3 nodes are management nodes.

On only one of the nodes, I get an error message and associated delay whenever I use a kubectl command. It looks like this:

$ time kubectl get pods
I0324 03:49:44.270996  514696 request.go:665] Waited for 1.156689289s due to client-side throttling, not priority and fairness, request: GET:https://127.0.0.1:16443/apis/authentication.k8s.io/v1?timeout=32s
NAME                      READY   STATUS    RESTARTS      AGE
sbnweb-5f9d9b977f-lw7t9   1/1     Running   1 (10h ago)   3d3h
shell-6cfccdbd47-zd2tn    1/1     Running   0             6h39m

real    0m6.558s
user    0m0.414s
sys     0m0.170s

The error message always shows a different URL each time. I tried looking up the error code (I0324) and haven't found anything useful.

The other two nodes don't show this behavior. No error message and completes the request in less than a second.

I'm new to k8s so I am not sure how to diagnose this kind of problem. Any hints on what to look for would be greatly appreciated.

AlanObject
  • 9,613
  • 19
  • 86
  • 142

2 Answers2

12

Here's a good write-up about the issue. For some cases rm -rf ~/.kube/cache will remove the issue.

gohm'c
  • 13,492
  • 1
  • 9
  • 16
  • 1
    So if I understand the issue correctly, the error message is more or less a natural result of the architecture of k8s, and there is no consensus on what a fix would look like. So for now, the work-around is just to ignore the message. – AlanObject Mar 24 '22 at 16:14
  • Can you try to resume the explanation here? Because this blog could not exist in the future :D – Hector Vido Aug 08 '22 at 18:19
  • 2
    For some cases this fix could be correct. For my case it hasnt brougth any change of the behaviour. The warning still exists. `12:41:04.334468 4784 request.go:665] Waited for 1.161945s due to client-side throttling, not priority and fairness, request: GET...` – snukone Oct 28 '22 at 10:46
0

I had a same error with kubectl on Windows. Deleting "http-cache" folder in ".kube" fixed it problem. c:\Users****.kube\http-cache\

Oleg
  • 1
  • 1
    Your answer could be improved with additional supporting information. Please [edit] to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers [in the help center](/help/how-to-answer). – Community Oct 26 '22 at 05:56