-
Notifications
You must be signed in to change notification settings - Fork 40.3k
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
Node E2E Test timeout #63240
Comments
@deads2k Is this expected? |
Since the namespace controller is using a single rest client rather than constructing lots of separate ones now, a commensurate increase in the allowed QPS should be made |
opened #63251 |
also, just noticed the node-e2e was still using default QPS of 5 (burst 10), which would have made it way slower than the normal namespace controller that runs |
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>. Bump QPS on namespace controller #62913 switched from using a client pool, where each groupVersionResource got its own rest client, to a single client. This increases the QPS to account for increased requests using a single rest client rate limiter. Fixes #63240 ```release-note NONE ```
Make sense. @liggitt Thanks for fixing this! I'll keep an eye on the test dashboard. |
I believe this fixes the node e2e test. Thanks! @liggitt |
Node E2E Test constantly timeout now.
https://k8s-testgrid.appspot.com/sig-node-kubelet#kubelet&graph-metrics=test-duration-minutes
From the test duration graph, it is very clear that the node e2e test duration changed from ~45m to >1h after the following 2 PRs are merged:
Since #63142 is only a README change, I believe it is #62913.
Because the test timeout, we don't know how long the test can be, or whether some component just stuck.
In node e2e test, we do use namespace controller. So I believe #62913 makes namespace controller work much slower or probably sometimes stuck.
This seems to be a significant regression to me. @deads2k @liggitt
/cc @kubernetes/sig-node-bugs @kubernetes/sig-api-machinery-bugs
The text was updated successfully, but these errors were encountered: