[release-4.21] SPLAT-2866: Add vSphere multi vCenter Day 2 - #3036
Conversation
(cherry picked from commit 76aebe7)
(cherry picked from commit 727806d)
(cherry picked from commit 43def11)
…e, and other review suggestions. (cherry picked from commit 7b21e9d)
(cherry picked from commit af78729)
(cherry picked from commit 291a26c)
…2 feature gate to TP"" (cherry picked from commit 1bf28d7)
(cherry picked from commit db1ed9c)
(cherry picked from commit 5c06b67)
The cherry-picked commits (openshift#2783, openshift#2886) used the enable(inTechPreviewNoUpgrade(), inDevPreviewNoUpgrade()) builder API added later on master; release-4.21 still uses enableIn(configv1.FeatureSet...), so the gate failed to compile. Switched to the equivalent enableIn() call and regenerated derived artifacts (features.md, openapi) to match.
|
Pipeline controller notification For optional jobs, comment This repository is configured in: LGTM mode |
openshift-ci-robot
added
the
jira/valid-reference
|
@vr4manta: This pull request references SPLAT-2866 which is a valid jira issue. DetailsIn response to this:
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 openshift-eng/jira-lifecycle-plugin repository. |
|
Hello @vr4manta! Some important instructions when contributing to openshift/api: |
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Repository: openshift/coderabbit/.coderabbit.yaml Review profile: CHILL Plan: Enterprise Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Comment |
openshift-ci
Bot
added
the
size/XXL
|
/pipeline required |
|
Scheduling tests matching the |
|
/test e2e-aws-ovn |
|
/override-sticky ci/prow/e2e-vsphere-ovn-techpreview Automated triage: This failure appears unrelated to the PR changes. Job classification: Eligible long-running e2e/integration job: the presubmit uses the Revision check: run Execution status: Tests executed. Cluster installation completed, the test phase ran for 1h44m46s, and the suite recorded 1,964 passes, 1 blocking failure, 1 flaky result, and 2,117 skips. The sole blocking failure was the exact test Completed supporting jobs: Fleet-wide failure rate: For this exact test outside this PR, the last 14 days show 16,242 passes and 12 hard failures out of 16,256 runs (99.91% pass rate), with 0 recorded flakes. All 12 failures share the same CRD cleanup Overlap assessment: The PR adds the Missing-coverage risk: Low for this failure: the vSphere e2e job completed its test phase, the relevant vSphere e2e job passed, and the only blocking result was a rare teardown timeout after the functional conversion assertions succeeded. Pending jobs were not counted as coverage. Rationale: The failure is a rare, environment-sensitive CRD deletion timeout, not a failed assertion in the PR's vSphere API/feature-gate changes. The current-revision and executed-test gates pass, and fleet evidence supports treating this as unrelated. If you disagree with this assessment, rerun the current job with AI-generated. Review for accuracy. |
|
@redhat-chai-bot: Overrode contexts on behalf of redhat-chai-bot: ci/prow/e2e-vsphere-ovn-techpreview These overrides will persist across retests on the current HEAD SHA. Pushing a new commit will clear them. Use DetailsIn response to this:
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. |
|
@vr4manta Do you know why the verify check is failing here? if we can get the failing tests aside from the crdify check this should be good to go. EDIT: Was finally able to see the results - looks like a codegen issue. |
Let me check. My guess is a fg that is present in this release but not master. |
|
/retest |
|
/pipeline required |
|
Scheduling tests matching the |
|
/override-sticky ci/prow/e2e-vsphere-ovn-techpreview Automated triage: This failure appears unrelated to the PR changes. Job classification: Eligible long-running e2e/integration job. The presubmit uses the Revision check: incoming/event SHA Execution status: Tests executed. Cluster installation completed and the e2e test phase ran. The run reported 1 blocking failure, 0 informing failures, 1,964 passes, 1 flaky result, and 2,117 skips. The exact blocking failure was Completed supporting jobs: Positive completed signal: Fleet-wide failure rate: The exact presubmit has 0% pass rate in the recent sample because both recorded runs belong to this PR. The equivalent 4.21 periodic job was 75.0% over the last 30 days (3 successes, 1 failure). For the exact test, the pass rate was 99.4% globally (5,533/5,564), 98.8% on 4.21 (242/245), and 100% on vSphere (373/373 globally in the recent window; 16/16 in the 4.21 variant view). The observed failures are timeout-related, including Open regressions: None found for this exact test in the relevant current 4.21 view. The direct Component Readiness bug association, OCPBUGS-73796, is Linked bugs: OCPBUGS-73796 is the only direct Sippy Overlap assessment: The PR changes the Missing-coverage risk: Low for this failure. The vSphere e2e job completed its test phase, the relevant non-TechPreview vSphere e2e job passed, and the only blocking result was a rare timeout in a generic ResourceQuota lifecycle test. This override does not waive the independently failed Prior bot activity on this SHA: No prior override on SHA Rationale: This is a clearly eligible, executed long-running vSphere e2e failure. The current revision matches, the functional e2e coverage ran, the failing test has a high fleet-wide pass rate with a known timeout signature and 100% recent vSphere pass rate, the only directly linked bug is closed, and the PR has no direct overlap with the failed ResourceQuota lifecycle test. The failure is therefore unrelated to the PR changes. If you disagree with this assessment, rerun the current job with AI-generated. Review for accuracy. |
|
@redhat-chai-bot: Overrode contexts on behalf of redhat-chai-bot: ci/prow/e2e-vsphere-ovn-techpreview These overrides will persist across retests on the current HEAD SHA. Pushing a new commit will clear them. Use DetailsIn response to this:
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. |
4a892dc to
93c7014
Compare
|
/pipeline required |
|
Scheduling tests matching the |
everettraven
left a comment
everettraven
left a comment
There was a problem hiding this comment.
/approve
/lgtm
/label backport-risk-assessed
openshift-ci
Bot
added
the
backport-risk-assessed
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: everettraven 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
|
Expected failure and tightened validation. /override ci/prow/verify-crdify |
|
@everettraven: Overrode contexts on behalf of everettraven: ci/prow/verify-crdify DetailsIn response to this:
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. |
|
/override-sticky ci/prow/e2e-gcp Automated triage: This failure appears unrelated to the PR changes. Job classification: Eligible long-running GCP end-to-end presubmit. The job is Revision check: Run SHA Execution status: Tests executed. The cluster was provisioned and
Completed supporting jobs: Fleet-wide failure rate: The exact job was 2/3 passing (66.7%) in the last 14 days; this small sample includes the current run. The exact test was 5464/5549 passing globally (98.5%) and 2466/2545 passing on GCP (96.9%) in the last 14 days. Sippy's 4.21 view was 56/57 passing globally (98.2%) and 22/23 passing on GCP (95.7%); the prior period was 100% over 205 runs. The failure is strongly GCP-concentrated. Open regressions: None found for the exact different-nodes test. The available 4.21 data shows a recent GCP-specific failure signal, but no open exact-test Component Readiness card was found. Linked bugs: Overlap assessment: The PR changes vSphere multi-vCenter Day 2 API/schema behavior, feature-gated manifests, generated CRDs, and API tests. It does not change GCP networking, LoadBalancer behavior, OVN, or the failing test. No plausible direct or indirect overlap with the failing GCP networking surface was found. Missing-coverage risk: Low for this failure. The failed test executed and is a known fleet-wide, GCP-concentrated failure; the PR's tested change is vSphere-specific. Other platform and upgrade end-to-end checks passed. Prior bot activity on this SHA: Rationale: The failure is a single GCP LoadBalancer UDP test with 85 failures in 5549 fleet runs and 79 failures in 2545 GCP runs, while the PR is confined to vSphere multi-vCenter API changes. The job meets the override gates, so the GCP failure is overridden. If you disagree with this assessment, rerun the current job with AI-generated. Review for accuracy. |
|
@redhat-chai-bot: Overrode contexts on behalf of redhat-chai-bot: ci/prow/e2e-gcp These overrides will persist across retests on the current HEAD SHA. Pushing a new commit will clear them. Use DetailsIn response to this:
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. |
|
@vr4manta: 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-sigs/prow repository. I understand the commands that are listed here. |
|
/verified by CI |
openshift-ci-robot
added
the
verified
|
@vr4manta: This PR has been marked as verified by DetailsIn response to this:
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 openshift-eng/jira-lifecycle-plugin repository. |
|
This forms part of an approved feature backport |
openshift-ci
Bot
added
the
jira/valid-bug
eb96322
into
openshift:release-4.21
6 participants
SPLAT-2866
Changes
Notes
This PR is attempting to backport the following PRs: