Repository navigation
ci!: parallelize validation and promote tested image candidates - #553
Merged
Merged
Conversation
Super-linter summary
All files and directories linted successfully For more information, see the GitHub Actions workflow run Powered by Super-linter |
|
Code Coverage Report Coverage ResultsTest Results
178 additional suites are available in the native reports.
Coverage
Overall: 83.37% 🟢 |
neilime
marked this pull request as ready for review
November 6, 2026 13:11
neilime
force-pushed
the
ci/475-p1-local-reductions
branch
from
November 10, 2026 21:19
daa2ba4 to
9f35667
Compare
This was referenced Nov 11, 2026
Merged
Record the five-run comparison, local fault checks and worker controls. Complete the P1 checklist and preserve the remaining upstream work. BREAKING CHANGE: backend.waitForDatabase.image overrides are removed. Database readiness now uses the backend workload image; remove those BusyBox overrides when upgrading the chart.
Use the released Node, common, container and publish workflows. Preserve explicit report paths and the prepared release commit with its target-ref input. Align local lint and record the P2 adoption protocol in ADR 0023.
neilime
commented
Nov 17, 2026
neilime
commented
Nov 17, 2026
neilime
commented
Nov 17, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Build CI images and production candidates independently, validate both applications and the exact candidate digests with Helm, then promote consumer tags only after every required check succeeds. This removes duplicate compilation and serial waits while preserving reports, signatures, provenance and fail-closed publication.
Validate PR checksverifies the expected job set and allowed skips.P8 acceptance freezes runtime revision
f30bb34. Five consecutive full CI samples, after a separate primer, all pass:Final elapsed range is 6m 21s-8m 33s. This is a historical comparison, not an isolated causal estimate: dependencies, tooling and runner images changed between P0 and P8. The ADR retains controlled phase comparisons and all slow/unsuccessful samples. Summed execution is not billing.
All 235 application build instructions across the twenty warm builds hit the cache. JUnit/LCOV preserve 1,178 passing frontend tests, 1,230 passing backend tests and 60 existing backend skips, with identical application coverage denominators. The 40 relocated routing cases match the former backend cases. All ten production digest handoffs match attestation, Helm installation and promotion. A real dependency-bot distribution update is verified in Publish #516.
Isolated cold and final-chart validation passes all 31 executed jobs. All four application builds disable layer imports/exports and mount persistence; all 47 build instructions execute without cache hits. Test inventories/coverage denominators match warm execution. The extra final-chart job validates release versions and promoted image defaults, uploads the exact tested archive, and passes checksum/content inspection. Both clusters and prepared archives are cleaned. It publishes no chart or GitHub release.
Review the P8 evidence, checkbox plan and current recovery guidance. This phase changes documentation only; raw evidence and the temporary validation workflow stay outside this PR.
Rollout remains open: successful main CI and bot publication/cancellation checks, required-check enforcement, and the next legitimate Pleey release. Generated-doc commits currently use
[skip ci]; requiringValidate PR checksneeds an approved bot bypass or a change to the bot workflow. The ADR proposes a main-only rule with a PR-only exemption for the existing bot app, explains the trade-off and leaves repository rules unchanged. No Pleey release is published for benchmarking.The final validation run passed in 16m 03s (30m 38s summed job time) and contains a retained setup outlier:
docker/setup-buildx-actiontook 600.6s, including a 579.3s pause after printing its BuildKit version. The subsequent backend build hit all 15 cached application instructions; all four builds hit all 47 instructions. The cause remains unestablished; this is outside the predefined five-attempt benchmark and is not discarded or presented as a layer-cache miss.Local validation: Markdown formatting/lint, 128 local links/anchors and
git diff --checkpass. Native CI for documentation commitaadc07apasses all 32 executed jobs with 28 expected nested skips. Runtime files match the measured revision.Refs #475; this PR does not close the issue.