fix(config): bump E2E CI mgmt cluster min node count (AROSLSRE-2032) - #6833
openshift-merge-bot[bot] merged 1 commit into
Conversation
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Raise mgmt.aks.userAgentPool.minCount from 4 to 5 for the ci00 and ci01 dev CI environments used by batch E2E runs. Pods land on more nodes, reducing the AKS infra alerts that gate batch E2E jobs on the shared ci01 environment.
There was a problem hiding this comment.
🟢 Approval recommended
The change is a minimal, consistent config bump applied in both source and rendered outputs with no apparent policy or schema impact.
Pull request overview
This PR adjusts the baseline AKS management-cluster user node capacity for the dev CI environments to reduce infra alert noise that can gate otherwise-successful parallel E2E runs. It updates the source config and the corresponding rendered artifacts for ci00 and ci01 in centralus.
Changes:
- Bump
mgmt.aks.userAgentPool.minCountfrom4to5for dev CIci00andci01(centralus). - Re-materialize rendered config outputs for both environments to reflect the new minimum.
File summaries
| File | Description |
|---|---|
| config/config.yaml | Raises mgmt.aks.userAgentPool.minCount to 5 for ci00 and ci01 in the source templated config. |
| config/rendered/dev/ci00/centralus.yaml | Updates rendered ci00 centralus config to match the new minCount: 5. |
| config/rendered/dev/ci01/centralus.yaml | Updates rendered ci01 centralus config to match the new minCount: 5. |
Review details
- Files reviewed: 3/3 changed files
- Comments generated: 0
- Review effort level: Lite
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: avollmer-redhat, raelga 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 |
|
/retest |
|
/hold Revision 8a23045 was retested 3 times: holding |
|
/retest |
|
/test e2e-parallel |
|
/unhold |
|
/test e2e-parallel |
What
Bumps
mgmt.aks.userAgentPool.minCountfrom 4 to 5 for theci00andci01dev CI environments (centralus), and re-materializes the rendered configs.Why
Batch E2E runs have been gated by AKS management-cluster infrastructure alerts on the shared ci01 environment, for example pull-ci-Azure-ARO-HCP-main-e2e-parallel #2095824620140630016, even though the E2E test suite itself passed. Running many HCP E2E tests in parallel on a small pool of nodes causes pod density issues on the short-lived management clusters. Raising the floor spreads pods across more nodes and gives us better pod distribution, reducing the impact of the alerts on CI success rate. This is a temporary mitigation while pod churn reduction and E2E test merging work lands.
Testing
Ran
cd config && make materializeand confirmed only the expecteduserAgentPool.minCountdiff inconfig/rendered/dev/ci00/centralus.yamlandconfig/rendered/dev/ci01/centralus.yaml.Special notes for your reviewer
Verified against the live
ARO HCP E2E Infrastructuresubscription that theci01environment'suserAgentPoolconfig (min 4/max 14, 3 pools viapoolCount) matches the deployeduserswft1/2/3node pools on a runningci01-*-mgmt-*cluster, confirming this is the right environment to change.PR Checklist