Running kubectl get all returns Throttling request errors
How can I debug and fix this issue?
I0223 10:28:04.717522 44883 request.go:655] Throttling request took 1.1688991s, request: GET:https://192.168.64.2:8443/apis/apps/v1?timeout=32s
I0223 10:28:14.913541 44883 request.go:655] Throttling request took 5.79656704s, request: GET:https://192.168.64.2:8443/apis/authorization.k8s.io/v1?timeout=32s
I0223 10:28:24.914386 44883 request.go:655] Throttling request took 7.394979677s, request: GET:https://192.168.64.2:8443/apis/cert-manager.io/v1alpha2?timeout=32s
I0223 10:28:35.513643 44883 request.go:655] Throttling request took 1.196992376s, request: GET:https://192.168.64.2:8443/api/v1?timeout=32s
I0223 10:28:45.516586 44883 request.go:655] Throttling request took 2.79962307s, request: GET:https://192.168.64.2:8443/apis/rbac.authorization.k8s.io/v1?timeout=32s
I0223 10:28:55.716699 44883 request.go:655] Throttling request took 4.600430975s, request: GET:https://192.168.64.2:8443/apis/node.k8s.io/v1beta1?timeout=32s
I0223 10:29:05.717707 44883 request.go:655] Throttling request took 6.196503125s, request: GET:https://192.168.64.2:8443/apis/storage.k8s.io/v1?timeout=32s
I0223 10:29:15.914744 44883 request.go:655] Throttling request took 7.99827047s, request: GET:https://192.168.64.2:8443/apis/acme.cert-manager.io/v1alpha2?timeout=32s
To diagnose kubectl commands, you select a level of verbosity when running the command.
If you run kubectl -v=9 you'll get a load of debug output.
If you look in there you may find the permissions for your cache directory within .kube are invalid.
I0511 09:28:13.431116 260204 cached_discovery.go:87] failed to write cache to /home/$USER/.kube/cache/discovery/CLUSTER_NAME/security.istio.io/v1beta1/serverresources.json due to mkdir /home/$USER/.kube/cache/discovery: permission denied
To resolve this I simply set the permissions to allow the cache data to be written.
chmod 755 -R ~/.kube/cache
This resolved the problem for me - hope it helps others.
According to Red Hat "Due to increasing number of Custom Resource Definitions (CRDs) installed in the RHOCP - Cluster the requests reaching for API discovery were limited by the client code."
I had a lot of CRDs on my OpenShift cluster and I observed this issue. Is it the case on your Kubernetes cluster?
Updating kubectl version worked for me.
just remove the sudo rm -rf ~/.kube/cache/
it's worked for me. don't spend too much time on this