-
Notifications
You must be signed in to change notification settings - Fork 15
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
✨ Bump to v1.17.2 #47
Conversation
sttts
commented
Feb 29, 2024
Bumps [k8s.io/apiserver](https://github.com/kubernetes/apiserver) from 0.28.0 to 0.28.1. - [Commits](kubernetes/apiserver@v0.28.0...v0.28.1) --- updated-dependencies: - dependency-name: k8s.io/apiserver dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <[email protected]>
Bumps [github.com/onsi/ginkgo/v2](https://github.com/onsi/ginkgo) from 2.11.0 to 2.12.0. - [Release notes](https://github.com/onsi/ginkgo/releases) - [Changelog](https://github.com/onsi/ginkgo/blob/master/CHANGELOG.md) - [Commits](onsi/ginkgo@v2.11.0...v2.12.0) --- updated-dependencies: - dependency-name: github.com/onsi/ginkgo/v2 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <[email protected]>
This PR shows how Gatekeeper has forked controller runtime to support the dynamic addition/removal of informers. Happy to flesh this out if people are interested. Not sure what the correct licensing actions are for moving code across CNCF projects. Fix lint error Signed-off-by: Max Smythe <[email protected]> Move mergeChan; add functional opts Signed-off-by: Max Smythe <[email protected]> Refactor functional parameters Signed-off-by: Max Smythe <[email protected]> Fix tests; refactor options type visibility Signed-off-by: Max Smythe <[email protected]> More interface changes Signed-off-by: Max Smythe <[email protected]> Fix lint Signed-off-by: Max Smythe <[email protected]> Add ginkgo so test flags are recognized Signed-off-by: Max Smythe <[email protected]> Split out non-blocking-get options; block on cancel Signed-off-by: Max Smythe <[email protected]> Add tests for removing informers Signed-off-by: Max Smythe <[email protected]> Fix missing newline Signed-off-by: Max Smythe <[email protected]> Oops, no newline Signed-off-by: Max Smythe <[email protected]>
…ot/go_modules/github.com/onsi/ginkgo/v2-2.12.0 🌱 Bump github.com/onsi/ginkgo/v2 from 2.11.0 to 2.12.0
…ot/go_modules/k8s.io/apiserver-0.28.1 🌱 Bump k8s.io/apiserver from 0.28.0 to 0.28.1
…tion option is true Signed-off-by: Troy Connor <[email protected]>
…pes-to-leader-election 🐛Add corev1, coordinationv1 scheme for leader election when LeaderElection manager option is true
Since v0.16.0 (kubernetes-sigs#2407), there is an import for all users of various opentelemetry libraries. This is caused by manager test -> pkg/metrics/filters -> k8s.io/apiserver -> otel-go. The issue is API server users a otel-go library from >1 year ago. This makes it impossible to import a modern otel-go library and a modern controller-runtime library together. Go is unable to prune the dependency *even though its only used in tests* (TBH, this surprised me!). By moving the tests that use `filters` under the `filters` package, though, this drops the required dependency on otel-go. Users that import the `filters` package will use it, of course, but everyone does not need to. So basically this refactors tests, but has user facing changes -- fewer dependencies are required to import controller-runtime core, allowing using the newer controller-runtime with newer otel-go. In parallel I will attempt to get k/k to update their otel-go version, but this will take months to ship at the earliest. IMO, this is worth fixing in the meantime as a v0.16.x patch.
Signed-off-by: Troy Connor <[email protected]>
🐛 Refactor tests to drop hard otel dependency
…-election-httpClient 🐛 Use http client from leaderElectionConfig
Bumps [actions/checkout](https://github.com/actions/checkout) from 3 to 4. - [Release notes](https://github.com/actions/checkout/releases) - [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md) - [Commits](actions/checkout@v3...v4) --- updated-dependencies: - dependency-name: actions/checkout dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <[email protected]>
Bumps [golang.org/x/sys](https://github.com/golang/sys) from 0.11.0 to 0.12.0. - [Commits](golang/sys@v0.11.0...v0.12.0) --- updated-dependencies: - dependency-name: golang.org/x/sys dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <[email protected]>
…ot/github_actions/actions/checkout-4 🌱 Bump actions/checkout from 3 to 4
…ot/go_modules/golang.org/x/sys-0.12.0 🌱 Bump golang.org/x/sys from 0.11.0 to 0.12.0
* Default namespace only for namespaced object * Add test
…bernetes-sigs#2479) * update subresource client to not incude metadata Signed-off-by: Troy Connor <[email protected]> * Test showing labels and annotations stored during fake client.Status().Update() * add test case to show it will not override other status fields when updating Signed-off-by: Troy Connor <[email protected]> * delete copyNonStatusFrom function, change test case description Signed-off-by: Troy Connor <[email protected]> --------- Signed-off-by: Troy Connor <[email protected]> Co-authored-by: Adam Berlin <[email protected]>
Before this, Update and Patch requests will override to-level status fields for objects that have a status resource if the status was initially empty. This happens because the code to set the status to what we already have in the unupdated object converts the object into a `map[string]any` using json serialization/deserialization, copies the original `status` field, then converts back. The json deserializer unfortunatelly does not properly clear the target, allowing data to remain that is absent in the json serialization.
…eroing 🐛 Fix status subresource getting updated on Update when it is empty
* Draft: Test that an object is updatable after updating its status. * Draft: ensure we udpate the new obj's accessor's ResourceVersion after doing the deep copy * Rework code to pass object back on status update --------- Co-authored-by: Adam Berlin <[email protected]>
🐛 Return NoResourceMatchError when appropriate for backwards compatibility.
Signed-off-by: Troy Connor <[email protected]>
…move-owner-reference-function 🌱 Create `RemoveOwnerReference` function in controllerutil to remove owner references from objects
📖 fix a grammer mistake in comment
Signed-off-by: Stefan Büringer [email protected]
🌱 Avoid dependency on apiserver/options if metrics/filters are used
…ckets ✨ Add 100/1000s buckets for prometheus workqueue histograms
Signed-off-by: Stefan Büringer [email protected]
🌱 Bump to controller-tools v0.14.0
📖 Fix Client Read/Write Functionality Comments
This change bumps https://github.com/evanphx/json-patch in order to pick up a bugfix that made it truncate large int64s because it internally treated them as float64s.
🐛 Don't truncate large int64s in MergePatch
Bumps [github.com/onsi/ginkgo/v2](https://github.com/onsi/ginkgo) from 2.13.2 to 2.14.0. - [Release notes](https://github.com/onsi/ginkgo/releases) - [Changelog](https://github.com/onsi/ginkgo/blob/master/CHANGELOG.md) - [Commits](onsi/ginkgo@v2.13.2...v2.14.0) --- updated-dependencies: - dependency-name: github.com/onsi/ginkgo/v2 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <[email protected]>
Signed-off-by: Joe Lanford <[email protected]>
🌱 bump go-apidiff to v0.8.2
…ot/go_modules/github.com/onsi/ginkgo/v2-2.14.0 🌱 Bump github.com/onsi/ginkgo/v2 from 2.13.2 to 2.14.0
This avoids: - Extra calls to https://host/apis/<group>/<version> when a version seen before and cached in apiGroups is deleted or marked as not served. - Returnning a valid mapping for a cached version that is deleted or not served anymore.
…ot/cherry-pick-2663-to-release-0.17 [release-0.17] 🐛 Clean restmapper cache if a version is notFound
…ge patch The fakeclient currently considers an apply patch to be a strategic merge patch. This is completely wrong, those are different things and apply patches are not supported in the fakeclientl, because in order to support them it would need the entire SSA logic which isn't implemented in upstream yet.
…ot/cherry-pick-2679-to-release-0.17 [release-0.17] ⚠ 🐛 Fakeclient: Do not consider an apply patch to be a strategic merge patch
When a group version is not found, if the group version is cached in apiGroups but not cached in knownGroups, the cache is not invalidated. Moreover and even worse, in that scenario an error is returned.
…ot/cherry-pick-2687-to-release-0.17 [release-0.17] 🐛 Fix lazy rest mapper cache invalidation
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
@sttts: The following test failed, say
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/test-infra repository. I understand the commands that are listed here. |