Add feature flag treat-pod-as-always-schedulable
The feature flag allows to declare that Pods in the system will eventually all get scheduled and Revisions should therefore not be marked unschedulable
Fixes #14862
Proposed Changes
Unfortunately, I never got feedback in above issue, so trying with a PR now:
- New feature
treat-pod-as-always-schedulableis added with default Disabled. - If enabled, the revision reconciler will skip the check whether a pod is not scheduled and therefore never mark a revision as unschedulable.
Release Note
You can now set the new feature `treat-pod-as-always-schedulable` to enabled in the config-defaults ConfigMap. As a result, Knative will not mark revisions as Unschedulable when a Pod is not scheduled. This makes sense if you want to omit this transient state in clusters that have cluster-autoscaling set up, and you can guarantee that all Pods will be eventually scheduled.
Hi @SaschaSchwarze0. Thanks for your PR.
I'm waiting for a knative member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.
Once the patch is verified, the new status will be reflected by the ok-to-test label.
I understand the commands that are listed here.
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.
Hm, I debugged the failing unit test and can see that when the code checks the config, that there is some config in the context with the right value but somehow what it gets has it empty. Need to check ...
@SaschaSchwarze0 gentle ping.
/ok-to-test
[APPROVALNOTIFIER] This PR is NOT APPROVED
This pull-request has been approved by: SaschaSchwarze0 Once this PR has been reviewed and has the lgtm label, please assign dprotaso for approval. For more information see the Code Review Process.
The full list of commands accepted by this bot can be found here.
Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment
@SaschaSchwarze0 gentle ping.
Rebased, but I still do not understand the unit test case failure in pkg/reconciler/revision/table_test.go. In particular, the custom context with the custom feature flag does not work = the code does not get the expected feature flag status and I do not understand why.
EDIT1: I think I now understand it. table_test.go does the following which overwrites all config customizations done on the Ctx of the row.
return revisionreconciler.NewReconciler(ctx, logging.FromContext(ctx), servingclient.Get(ctx),
listers.GetRevisionLister(), controller.GetEventRecorder(ctx), r,
controller.Options{
ConfigStore: &testConfigStore{
config: reconcilerTestConfig(),
},
})
EDIT2: Though, I do not understand why it works for other unit tests then (like the one that sets PodSpecInitContainers to Enabled).
EDIT3: Maybe because only the controller uses the configstore from its options.
EDIT4: Fixed.
Codecov Report
:white_check_mark: All modified and coverable lines are covered by tests.
:white_check_mark: Project coverage is 80.84%. Comparing base (19b9a09) to head (5303dc0).
:warning: Report is 213 commits behind head on main.
Additional details and impacted files
@@ Coverage Diff @@
## main #15397 +/- ##
==========================================
+ Coverage 80.78% 80.84% +0.06%
==========================================
Files 222 222
Lines 18025 18029 +4
==========================================
+ Hits 14561 14576 +15
+ Misses 3092 3084 -8
+ Partials 372 369 -3
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
:rocket: New features to boost your workflow:
- :snowflake: Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
PR needs rebase.
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.
This Pull Request is stale because it has been open for 90 days with
no activity. It will automatically close after 30 more days of
inactivity. Reopen with /reopen. Mark as fresh by adding the
comment /remove-lifecycle stale.
Hm, I missed that this PR was auto-closed. Let me try if I can revive it or need to open a new one.
/reopen
@SaschaSchwarze0: Failed to re-open PR: state cannot be changed. The sascha-treat-pod-as-always-schedulable branch was force-pushed or recreated.
In response to this:
Hm, I missed that this PR was auto-closed. Let me try if I can revive it or need to open a new one.
/reopen
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.