-
Notifications
You must be signed in to change notification settings - Fork 435
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
/openapi/v3 api endpoint takes too long to respond #1588
Comments
Hey @L1ghtman2k ! Thanks for creating this issue! Does disabling the proxy metrics server help? Also is it faster within a smaller vCluster? |
@FabianKramm. I should also mention that our vcluster runs on top of microk8s |
Could this be the cause for #1589 as well? |
@L1ghtman2k thanks for the info, with what helm values are you deploying vClsuter with and without metrics server? |
@L1ghtman2k Following the request from Fabian above, which backing store are you using for your virtual cluster? The amount of CRDs and potential expensive discovery requests resulting from this might strain the API server and its underlying backing store (See: #1589 (comment)) |
Quick heads up, I think there a changes in |
Thanks, I will take a look when we do the upgrade |
What happened?
kubectl get --raw /openapi/v3
takes > 30 seconds to respond. This behavior is inconsistent, as sometimes it is instantaneous, and other times it takes too longThis seemed to be caused by a large number of CRDs on the cluster
Intern, terraform resources like helm_release end up timing out, causing errors like:
What did you expect to happen?
This is a multipart issue, that can be resolved either by fixing helm terraform provider, to allow larger timeouts.
I am creating this issue here because it seems to be in part related to vcluster only, as our EKS deployments are just fine. (On the other hand, we aren't running k8s 1.27 on our EKS clusters, which means we likely not leveraging /openapi/v3)
How can we reproduce it (as minimally and precisely as possible)?
Install vcluster with lots of CRDs (monitoring stack, kyverno), size may matter, here are our CRDs:
run
kubectl get --raw /openapi/v3
a few times, to see how long it takes to respondAnything else we need to know?
No response
Host cluster Kubernetes version
Host cluster Kubernetes distribution
Client Version: v1.29.2
Kustomize Version: v5.0.4-0.20230601165947-6ce0bf390ce3
Server Version: v1.29.2
vlcuster version
vcluster version 0.19.4
Vcluster Kubernetes distribution(k3s(default)), k8s, k0s)
k3s
OS and Arch
amd64
The text was updated successfully, but these errors were encountered: