-
Notifications
You must be signed in to change notification settings - Fork 38.7k
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
reintroduce resourcequota.NewMonitor #120777
reintroduce resourcequota.NewMonitor #120777
Conversation
- this function is used by other packages and was mistakenly removed in 397cc73 - let resource quota controller use this constructor instead of an object instantiation
Thanks, but as #115103 (comment) , we do not guarantee compatibility of internal packages! Perhaps we can also look at @thockin's thoughts on the issue. |
It's not as easy as the issue claims, and there will be cases where we'll continue keeping those functions as is. |
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
/approve
LGTM label has been added. Git tree hash: e3f9cf7e529c2046a4da8eb10626ed27c30111d3
|
/triage accepted |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: atiratree, soltysh 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 |
/test pull-kubernetes-node-e2e-containerd |
@atiratree Do you have time to create the cherry-pick PR to v1.28 for this regression? |
…20777-upstream-release-1.28 Automated cherry pick of #120777: reintroduce resourcequota.NewMonitor
FWIW, this was not a regression. Stability of go functions / signatures in k8s.io/kubernetes aren't part of the API surface area we guarantee. |
Agree emphatically - we do not support people importing from within k8s.io/kubernetes and any project that depends on it may find themselves on the receiving end of such a cleanup. |
ack, thanks for removing the regression part |
* update last_rebase.sh * update changelog * update microshift/go.mod * update microshift/vendor * update etcd/go.mod * update etcd/vendor * update component images * update manifests * update buildfiles * add patch to reintroduce resourcequota.NewMonitor See kubernetes/kubernetes#120777 * apply resourcequota.NewMonitor patch * update location of ServiceIPRange The definition of ServiceIPRange changed in upstream kubernetes, so update the way we import it. * remove trailing whitespace in build_images.sh * USHIFT-1638: fix current release package source template At the start of a release cycle there are no EC or RC builds for the current release, so we should skip adding the package source. Package sources are skipped if their template produces an empty file when rendered. The crel template was producing a file with 1 blank line. Fix the whitespace marker at the end of the template to avoid an extra blank line. * USHIFT-1638: fix detection of empty blueprint file Use the right local variable for the file to check for content to detect that a template for a blueprint rendered empty. * when crd manager receives cancel notice return the error * USHIFT-1638: allow repeated configuration of logging Tell the logging code that it's OK to receive reconfiguration instructions unless those instructions are different. This overrides the default behavior of erroring out if any instructions are received a second time. That default behavior causes issues with embedded components that all use the same logging code and are not expecting to be compiled together into the same process, as they are in MicroShift. See comments in k8s.io/component-base/logs/api/v1/options.go for details. --------- Co-authored-by: ci-robot <[email protected]>
What type of PR is this?
/kind bug
What this PR does / why we need it:
Which issue(s) this PR fixes:
Fixes #
After bumping kubernetes dependencies, other projects depending on this public constructor will fail to compile. E.g. openshift/cluster-policy-controller#132. Since the QuotaMonitor members are private, there is no workaround.
Special notes for your reviewer:
This PR should be backported to 1.28 for compatibility reasons.
Does this PR introduce a user-facing change?
Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.: