-
Notifications
You must be signed in to change notification settings - Fork 39k
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
Revert kubelet to default to ttl cache secret/configmap behavior #74755
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/lgtm
/hold would like @kubernetes/sig-scalability-pr-reviews @kubernetes/sig-node-pr-reviews input |
05f48a5
to
cd1c9f9
Compare
/cc @yujuhong |
/lgtm |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: liggitt, yujuhong The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
/retest |
/hold cancel |
…5-upstream-release-1.12 Automated cherry pick of #74755: Revert kubelet to default to ttl cache secret/configmap
@liggitt, this PR seems to be related to recent flakiness of our scalability tests, see #75171 (comment) What is the long-term plan here? Are we going to revert this PR? |
not until our http/2 dependency is bumped to go1.12 levels and we scale test the watching secret manager (see #74781 and #71501 (comment)) |
I see, thanks. In that case we need to figure out what exactly is causing flakiness in #75171, there might be some other bug involved there. |
…5-upstream-release-1.13 Automated cherry pick of #74755: Revert kubelet to default to ttl cache secret/configmap
What type of PR is this?
/kind bug
What this PR does / why we need it:
Reverts the kubelet to pre-1.12 behavior for fetching configmaps/secrets, due to http/2 stream saturation issues (golang/go#27044, fixed in go1.12, but not available to k8s 1.12/1.13). Once we update to golang 1.12+, and have scale tests in place to test this scenerio more thoroughly, we can re-enable the watching manager by default.
Which issue(s) this PR fixes:
xref #74412
Does this PR introduce a user-facing change?:
/cc @wojtek-t
@kubernetes/sig-scalability-pr-reviews