4.10 downstream/upstream sync - second half#204
Closed
njhale wants to merge 29 commits intoopenshift:masterfrom
Closed
4.10 downstream/upstream sync - second half#204njhale wants to merge 29 commits intoopenshift:masterfrom
njhale wants to merge 29 commits intoopenshift:masterfrom
Conversation
…images (openshift#771) Signed-off-by: Joe Lanford <joe.lanford@gmail.com> Upstream-repository: operator-registry Upstream-commit: f5dcf40f01b8707194fdde07818c8b440ee010fa
…cle-manager. (#2337) * Remove resolver dependency on registry.CatalogKey. Signed-off-by: Ben Luddy <bluddy@redhat.com> * Remove hard resolver dependency on the gRPC registry client. The most significant change is the introduction of a new Source interface, which represents a single source of cache entries. In the existing implementation, the cache was hardcoded to fetch content from the operator-registry gRPC API on misses. This was a barrier to off-cluster tools and other applications that should be able to perform resolution without running a registry server. A number of test fixtures were affected (for the better) because it's now possible to directly configure the real Cache implementation to retrieve test content without using of test doubles. Signed-off-by: Ben Luddy <bluddy@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 9e4a778a44abb8705fb8ef95ad0bdaaebfa0ecfc
…338) Update the root go.mod and remove the replace pin for sigs.k8s.io/structured-merge-diff. It looks like the "// pinned because no tag supports 1.18 yet" message may have also been an artifact of removing the helm replace pin that was previously in place. Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 3eb0d7be413d7c185441cb3fc81dadb2b67706e7
* add timestamp field Signed-off-by: Ish Shah <ishah@redhat.com> * Add timestamp to scorecard test result Signed-off-by: Ish Shah <ishah@redhat.com> * Change type and struct Signed-off-by: Ish Shah <ishah@redhat.com> * Update pkg/apis/scorecard/v1alpha3/test_types.go Co-authored-by: Joe Lanford <joe.lanford@gmail.com> * make manifests/generate; Signed-off-by: Ish Shah <ishah@redhat.com> * Update pkg/apis/scorecard/v1alpha3/test_types.go Co-authored-by: Joe Lanford <joe.lanford@gmail.com> Co-authored-by: Joe Lanford <joe.lanford@gmail.com> Upstream-repository: api Upstream-commit: 7d591fff8bd551160e51dedef17debbe52145019
* Add validation for Kind and APIVersion Signed-off-by: Josef Karasek <jkarasek@redhat.com> * Update pkg/validation/internal/csv.go Co-authored-by: Eric Stroczynski <ericstroczynski@gmail.com> Co-authored-by: Eric Stroczynski <ericstroczynski@gmail.com> Upstream-repository: api Upstream-commit: c9b1e7ef82b8081a3eb52c48d0ca9039f2e3a27e
…r packages. (#2340) * Stop using the registry's Bundle proto type for resolution. The resolver component does not care whether a given cache entry was populated by use of the registry gRPC API, so it should ultimately be possible to drop the "Bundle" field from cache entries altogether. For now, catalog-operator's "step resolver" still depends on a Bundle proto field in order to unpack inline (as opposed to image-based) packages. Signed-off-by: Ben Luddy <bluddy@redhat.com> * Remove exported resolver cache entry creation funcs. NewOperatorForBundle and NewOperatorFromV1Alpha1CSV couple the resolver cache package to the registry gRPC API and the operators/v1alpha1 CSV API, respectively. They're effectively used only by the resolver package, so they can be moved there and unexported. The remaining calls to NewOperatorFromV1Alpha1CSV can be extracted from core resolver logic into a Source implementation that is responsible for the conversion from ClusterServiceVersion to cache.Snapshot entry, and the remaining call to NewOperatorFromBundle is used only by the Source implementation based on the registry gRPC API. Signed-off-by: Ben Luddy <bluddy@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: b849eb79c6fbb1c445ed3e9945e1668f6066ea6e
…ift#154) * modified the controller-runtime and controller-tools versions * ran vendor command and added files * ran make manifests Upstream-repository: api Upstream-commit: a80624eab36b2005fa5d84c9b539bb4817e44daa
* .github/workflows: Separate out the e2e jobs into workflows Update the .github/workflows/test-scripts.yml GH action workflow and move all the jobs listed there into separate workflows. This allows us to retest a single workflow at a time as a workaround to GH not allowing you to retest individual jobs in a workflow. Signed-off-by: timflannagan <timflannagan@gmail.com> * .github/workflows: Add the 'ci' workflow name to all e2e-related jobs Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 966ce43d5bde93848c45e1aaf4c6a5c110c37740
Add a badge that links to the olm-dev channel in the Kubernetes Slack instance. Signed-off-by: Nick Hale <njohnhale@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 94af95a7580f45f6a0b45bd12d5aa37649b74a47
Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 864b58ddc63742b53ecdf21f463e13ac2ce9de7e
…file (openshift#773) Signed-off-by: Eric Stroczynski <ericstroczynski@gmail.com> Upstream-repository: operator-registry Upstream-commit: 1c8557637731bbe28166d854ea8a16fd86c2995e
* .github/workflows: Don't run CI checks for markdown-only updates Update the ci-prefixed workflows and ensure that CI isn't running when only markdown files are updated. Signed-off-by: timflannagan <timflannagan@gmail.com> * .github/workflows: Use paths-ignore instead of paths for ignoring certain files/paths Signed-off-by: timflannagan <timflannagan@gmail.com> * .github/workflows: Remove the need to ignore the root doc/ path The '**/*.md' filter pattern [1] should ignore _all_ markdown extension files anywhere in the repository. [1] https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#patterns-to-match-file-paths Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 2fd014019ca8cc86ae859da797e814b72c299257
Signed-off-by: Eric Stroczynski <ericstroczynski@gmail.com> Upstream-repository: operator-registry Upstream-commit: a3253eb2fa97be52883af0a0c195d4a95500c04a
Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: e48737c5fc30c867d7f3ea3ce471c86050aa9a89
…#2370) Currently, OLM only caches configmap with olm-managed label to reduce memory usage footprint. There is an automatic mechanism to inject this label to all configmaps that belong to catalogsource. However, for e2e test, this label should be added by default to avoid extra logic to be executed. Signed-off-by: Vu Dinh <vudinh@outlook.com> Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 8b0ac13809b5155e4b7f1415e2fb37a060c3fe07
Update the root Makefile and update the `e2e` target to use the ginkgo executable instead of `go test ...` as logs produce when running the e2e suite aren't captured until after the process has completed. This leads to situations where failing tests aren't immediately communicated back and loss of logs entirely in the case where the process was manually teminated. Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 584e12a9819b8aea1205332c4eba72a46d609fad
Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 396b5461555c2498507b1fff20fccc3e7ae15420
Use server-side apply to update a fixture and avoid false negatives due caused by resource contention. Signed-off-by: Nick Hale <njohnhale@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 6b3935294cf09261d81b9233ee5fd181ae63e6e0
Signed-off-by: Daniel Sover <dsover@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 1a6a03ec9e18a5e4394b06a053258846c6d52872
Signed-off-by: Vu Dinh <vudinh@outlook.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: a03dd0c21020d1f2c0486e00fb5f6660f1d1e363
The PreExistingCRDOwnerIsReplaced test case is no longer applicable as the ownership replacement check has been removed from olm. This test was introduced when olm only allowed one single ownership of CRD. Signed-off-by: Vu Dinh <vudinh@outlook.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 1237d38e0661731f0b62d65dd3883f366452454e
- Add more config options to the e2e and e2e-local test options; e.g. E2E_TEST_SEED for reproducing test runs - Refactor e2e-local target to extend the e2e target Signed-off-by: Nick Hale <njohnhale@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 07dd3a00af7b580ac3d41ff0d51ef982ae019224
namespaces Update the root Makefile and extend the e2e/olm target to explicitly configure the ginkgo options that gets filtered to the staging/operator-lifecycle-manager e2e target when running on OpenShift clusters. Signed-off-by: timflannagan <timflannagan@gmail.com>
…ertions (#2403) * test/e2e: Wrap flake-prone garbage collection tests in eventually assertions Update the test/e2e/gc_e2e_test.go and wrap any error-prone test blocks in eventually assertions to reduce the number of flakes that occur when attempting to create/update/delete/etc. Kubernetes resources during individual CI runs on these garbage collection specs. Signed-off-by: timflannagan <timflannagan@gmail.com> * test/e2e: Replace context.TODO() with context.Background() Upstream-repository: operator-lifecycle-manager Upstream-commit: 23d84fa7dedff7546f1622253aabeaeee7ec693a
…ach block (#2387) Signed-off-by: Daniel Sover <dsover@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: a94b10b8cf7f0b2632eebdbd4fe85dd16520b89e
Signed-off-by: Daniel Sover <dsover@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 4c3c8f1b1bf6c64271f81a4e64043a081c9d73b7
…419) This commit introduces a change to the e2e tests that rely on a grpc catalogSource that is stored in an off cluster registry. These tests originally required that the following happen within a minute: 1. The catalogSource is installed and ready to serve. 2. The contents of the bundle are unpacked and placed in an installPlan. 3. The CSV is installed and in the ready state. The tests will now wait for the catalogSource to be ready to serve before moving onto steps 2 and 3. Signed-off-by: Alexander Greene <greene.al1991@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 700d282adbe09cc0b5d84d70411addb7400f9f79
Contributor
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: njhale The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-ci
Bot
added
the
approved
Contributor
Author
|
/retest |
Contributor
Author
|
/retest Looking at steady-state for |
Contributor
|
@njhale: The following tests failed, say
Full PR test history. Your PR dashboard. DetailsInstructions 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. |
Contributor
Author
|
superseded by #205 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
9 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Second half of commits in #196
Subset of: