gcp: migrate e2e-gcp-ovn-rt to N4 with hyperdisk-balanced (1/2) - #84090
Conversation
Forward migration (step 1) following the revert in openshift#84065. GCP N2 quota is very constrained, so move to N4 where the platform supports it. The openshift-e2e-gcp-ovn-rt workflow is a clean target: all of its consumers (e2e-gcp-ovn-rt-rhcos10-techpreview in nightly 4.22/4.23/5.0/5.1) run on releases >= 4.17, which is the floor for GCP N4 + hyperdisk-balanced support. N4 only supports hyperdisk-balanced disks (not pd-*), so this: - sets COMPUTE_NODE_TYPE=n4-standard-8 and COMPUTE_DISK_TYPE=hyperdisk-balanced on the workflow; - adds ipi-conf-gcp-osdisk-disktype (node OS disk) and storage-conf-storageclass-pre-set-default-gcp-hyperdisk (default StorageClass for dynamically provisioned PVCs, e.g. monitoring) before install, matching the openshift-e2e-gcp-csi-custom-worker convention; - drops the now-redundant COMPUTE_NODE_TYPE: n2-standard-8 override from the four consumers so node and disk type stay a single source of truth in the workflow. Depends on openshift#84065 (revert) merging first. Assisted-By: Claude Opus 4.8
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository YAML (base), Central YAML (inherited) Review profile: CHILL Plan: Enterprise Run ID: 📒 Files selected for processing (3)
Included review availability: Your plan provides up to 12 included reviews per hour; 10 remain after this review. WalkthroughThe GCP OVN realtime workflow adds OS-disk and HyperDisk settings and uses ChangesGCP OVN realtime workflow
Nightly configuration alignment
Estimated code review effort: 2 (Simple) | ~10 minutes Merge Risk: ⚪ Minimal · up to The workflow moves from N2 to N4 with hyperdisk-balanced storage and removes redundant consumer overrides; no actionable merge-blocking risk remains in the supplied current-head review evidence. Suggested reviewers: 🚥 Pre-merge checks | ✅ 15✅ Passed checks (15 passed)
Full details: Title checkExplanation The title clearly and concisely summarizes the main change: migrating the GCP OVN realtime workflow from N2 to N4 with hyperdisk-balanced storage. The step indicator is also consistent with the stated two-step migration. Full details: Docstring CoverageExplanation No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0 files. (3 skipped: 3 unsupported.) Full details: Stable And Deterministic Test NamesExplanation PASS. The PR changes only CI YAML. The workflow changes add pre-step references and set Full details: Test Structure And QualityExplanation PASS — The PR changes only five YAML configuration/workflow files. The exact two-commit diff contains no Ginkgo or other test-source files, so it introduces no It blocks, resource setup, waits, or assertions for this check to review. Full details: Microshift Test CompatibilityExplanation PASS — The pull request changes only CI YAML workflow and test configuration. The PR diff adds no Go files and no new Ginkgo constructs such as Full details: Single Node Openshift (Sno) Test CompatibilityExplanation PASS: The pull request changes only five YAML configuration/workflow files. The diff adds no Ginkgo test declarations or test source; it only changes worker, disk, pre-step, and TEST_SKIPS settings. Therefore, the SNO compatibility check is not applicable. Full details: Topology-Aware Scheduling CompatibilityExplanation PASS — The PR changes only CI workflow/configuration YAML. The diff changes GCP machine and disk environment variables, adds existing pre-steps, removes consumer overrides, and adds a test skip. It does not add or modify deployment manifests, operator code, or controllers, and it introduces none of the listed pod scheduling constraints such as affinity, topology spread, node selectors, tolerations, replica counts, or PDBs. Full details: Ote Binary Stdout ContractExplanation PASS: The PR changes only five YAML workflow/configuration files. The diff contains no Go files, OTE binary code, suite setup, process entry points, or stdout/logging writes. The added GCP settings and TEST_SKIPS values do not introduce an OTE stdout contract violation. Full details: Ipv6 And Disconnected Network Test CompatibilityExplanation PASS — The pull request changes only five YAML CI configuration files. The diff adds no Ginkgo test declarations, test source, IPv4 parsing, hardcoded test addresses, or external-network test logic. The new workflow entries configure GCP disks and nodes, and the consumer changes remove an override or add Full details: No-Weak-CryptoExplanation The PR diff only changes YAML workflow and test configuration. It switches GCP node and disk settings, adds two pre-step references, removes old node overrides, and adds a Persistent Disk test skip. No MD5, SHA1, DES, RC4, 3DES, Blowfish, ECB, cryptographic implementation, or secret/token comparison appears in the changed lines. Full details: Container-PrivilegesExplanation PASS: The pull-request diff changes only workflow steps, GCP disk settings, and test environment values. It adds no Full details: No-Sensitive-Data-In-LogsExplanation PASS: The PR changes only GCP machine, disk, workflow-step, and test-skip configuration. The changed lines contain no passwords, tokens, API keys, PII, session IDs, hostnames, or customer data. The two added pre-steps emit only configuration/status text; the os-disk step reads the non-sensitive ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
|
@mkowalski: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
/pj-rehearse |
|
/verified by CI |
|
@mkowalski: 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. |
|
@mkowalski: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
In 3 failed tests we created an N4 instance but it vanished a few seconds later. This may be an instance of our well-known GCP issue. Or not. We shall see. |
|
/pj-rehearse periodic-ci-openshift-release-main-nightly-4.22-e2e-gcp-ovn-rt-rhcos10-techpreview periodic-ci-openshift-release-main-nightly-5.0-e2e-gcp-ovn-rt-rhcos10-techpreview periodic-ci-openshift-release-main-nightly-5.1-e2e-gcp-ovn-rt-rhcos10-techpreview Re-running the three failed rehearsals to determine whether the N4 worker provisioning failure ( This comment was generated using AI. Please verify before acting on it. |
|
@mkowalski: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
/pj-rehearse periodic-ci-openshift-release-main-nightly-5.0-e2e-gcp-ovn-rt-rhcos10-techpreview The previous 5.0 rehearsal failed on a transient CI infra flake (gcloud auth token refresh could not resolve This comment was generated using AI. Please verify before acting on it. |
|
@mkowalski: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
Rehearsal root-cause summaryAll four A) Intermittent N4 worker provisioning (install fails) — transient
B) Systematic N4 storage-conformance incompatibility (e2e fails) — deterministicOn the runs that did install (4.23, 5.0), N4 supports only hyperdisk-balanced, not Persistent Disk (regional or zonal). The ConclusionThis job runs the full parallel conformance suite, which exercises GCE Persistent Disk — something N4 cannot do. So moving Recommendation: keep the GCP conformance jobs on N2 (they need PD‑capable workers for the storage conformance tests). N4 remains appropriate only where the suite does not exercise GCE‑PD (e.g. the CSI/hyperdisk‑specific jobs). I'll rescope the forward‑migration accordingly. This comment was generated using AI. Please verify before acting on it. |
The [sig-storage][Driver: pd.csi.storage.gke.io] "regional PD should store data and sync across zones" conformance test provisions a regional pd-standard PersistentDisk and attaches it to a worker. N4 supports only hyperdisk-balanced (not Persistent Disk), so the attach fails deterministically now that e2e-gcp-ovn-rt-rhcos10-techpreview runs on n4-standard-8: AttachVolume.Attach failed: googleapi: Error 400: Regional disks is not supported for n4-standard-8 machine type Skip that single test via TEST_SKIPS on the 4.23/5.0/5.1 consumers (the releases whose payload contains it; it is absent in <=4.22). The rest of openshift/conformance/parallel still runs, and N2 jobs keep the test. This is the tactical unblock. The durable fix makes the test skip itself on hyperdisk-only worker families in openshift/origin (test/extended/storage/gce_pd_regional.go), after which these TEST_SKIPS lines can be dropped. Assisted-By: Claude Opus 4.8
|
[REHEARSALNOTIFIER]
Interacting with pj-rehearseComment: Once you are satisfied with the results of the rehearsals, comment: |
|
/pj-rehearse PD test should be disabled for N4 now |
|
@mkowalski: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
It skips Once #31572 and its backports (release-5.1, release-5.0, release-4.23) reach the payloads, the This comment was generated using AI. Please verify before acting on it. |
|
/retest-required |
|
@mkowalski: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
I was confused as to why the test job wasn't skipped in 4.22 but it's new in 4.23/5.0. |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: mkowalski, petr-muller, sdodson 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 |
971298a
into
openshift:main
|
@mkowalski: The following test 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. |
What this PR changes
Moves the
openshift-e2e-gcp-ovn-rtworkflow from GCP N2 to N4 worker nodes. N4 only supportshyperdisk-balanceddisks (not Persistent Disk), so the workflow also switches the disk type and makeshyperdisk-balancedthe default StorageClass. All four consumers of this workflow run on OCP 4.22+, safely above the 4.17 floor for N4 + hyperdisk.1) Workflow
presteps — two steps added before installflowchart TB S1["ipi-conf-gcp"] --> S2["ovn-conf"] --> S3["ipi-conf-debug-kdump-configure-logs"] --> S4["rhcos-conf-osstream"] --> S5["ipi-conf-gcp-osdisk-disktype<br/>NEW - sets worker OS disk = hyperdisk-balanced"]:::new --> S6["storage-conf-storageclass-pre-set-default-gcp-hyperdisk<br/>NEW - makes hyperdisk-balanced the default StorageClass"]:::new --> S7["ipi-install-stableinitial"] classDef new fill:#d4f7d4,stroke:#2ea043,stroke-width:2px,color:#000;2) Workflow
env— machine type + disk typeflowchart LR O1["COMPUTE_NODE_TYPE = n2-standard-8"]:::old -->|changed| N1["COMPUTE_NODE_TYPE = n4-standard-8"]:::new O2["(no disk type set)"]:::old -->|added| N2["COMPUTE_DISK_TYPE = hyperdisk-balanced"]:::new classDef old fill:#ffecec,stroke:#d1242f,color:#000; classDef new fill:#d4f7d4,stroke:#2ea043,color:#000;3) The four consumer configs — drop the redundant override
Each
e2e-gcp-ovn-rt-rhcos10-techpreviewtest used to pin the old N2 type. Removing that override lets the test inherit the new N4 + hyperdisk settings straight from the workflow (single source of truth).flowchart TB C["nightly-4.22 / 4.23 / 5.0 / 5.1<br/>test: e2e-gcp-ovn-rt-rhcos10-techpreview"] C -->|before| CB["env: COMPUTE_NODE_TYPE = n2-standard-8<br/>(overrode the workflow)"]:::old C -->|after| CA["override removed → inherits<br/>n4-standard-8 + hyperdisk-balanced"]:::new classDef old fill:#ffecec,stroke:#d1242f,color:#000; classDef new fill:#d4f7d4,stroke:#2ea043,color:#000;Files changed (5)
step-registry/openshift/e2e/gcp/ovn-rt/openshift-e2e-gcp-ovn-rt-workflow.yamlhyperdisk-balanced; add osdisk-disktype + hyperdisk-StorageClass stepsconfig/.../openshift-release-main__nightly-4.22.yamlCOMPUTE_NODE_TYPE: n2-standard-8overrideconfig/.../openshift-release-main__nightly-4.23.yamlconfig/.../openshift-release-main__nightly-5.0.yamlconfig/.../openshift-release-main__nightly-5.1.yamlContext: forward step of the N2→N4 migration on top of the (merged) revert #84065.
This description was generated using AI. Please verify before acting on it.
Summary by CodeRabbit
openshift-e2e-gcp-ovn-rtworkflow from GCP N2 to N4.n4-standard-8workers withhyperdisk-balanceddisks.