Skip to content

ci: don't fail-fast the wheel matrix#1692

Closed
bdraco wants to merge 1 commit into
masterfrom
ci-wheels-no-fail-fast
Closed

ci: don't fail-fast the wheel matrix#1692
bdraco wants to merge 1 commit into
masterfrom
ci-wheels-no-fail-fast

Conversation

@bdraco
Copy link
Copy Markdown
Member

@bdraco bdraco commented May 16, 2026

Summary

Add fail-fast: false to the build_wheels job's matrix strategy so one arch's failure no longer cancels the other ~30 wheel builds.

Motivation

run/25974564348 had a single failing entry — Wheels for ubuntu-latest (musllinux) armv7l cp39 — and every other arch was cancelled mid-build. upload_pypi was skipped, so 0.149.0 landed on PyPI as sdist only. With fail-fast: false, the other arches would have completed and upload_pypi would have published whatever the matrix produced; a fix could then re-target just the broken arch.

fail-fast: false is already set on the test matrix at line 51 — this just brings build_wheels in line.

Test plan

  • Merge → next release's wheel matrix continues even if a single arch fails.
  • Sanity-check that upload_pypi (which depends on build_wheels) still gates correctly. With fail-fast: false, a single-arch failure means build_wheels is failure, so upload_pypi would still be skipped — same gating behaviour, just better diagnostics on which arch broke.

Note

This is a ci: change, so it's excluded from semantic-release's changelog (exclude_commit_patterns in pyproject.toml) and won't bump the version on its own.

A single arch's failure was cancelling every other wheel job (see
0.149.0, where one cp39 entry took down the entire build_wheels run
and left PyPI with sdist only). Setting `fail-fast: false` lets the
other ~30 jobs keep going, so a fix targets just the one broken arch.
@bdraco
Copy link
Copy Markdown
Member Author

bdraco commented May 16, 2026

nah, we want to fail .. better to fail than partial

@bdraco bdraco closed this May 16, 2026
@bdraco bdraco deleted the ci-wheels-no-fail-fast branch May 16, 2026 23:02
@codspeed-hq
Copy link
Copy Markdown

codspeed-hq Bot commented May 16, 2026

Merging this PR will not alter performance

✅ 6 untouched benchmarks


Comparing ci-wheels-no-fail-fast (18f056f) with master (591288b)

Open in CodSpeed

@codecov
Copy link
Copy Markdown

codecov Bot commented May 16, 2026

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.76%. Comparing base (591288b) to head (18f056f).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files
@@           Coverage Diff           @@
##           master    #1692   +/-   ##
=======================================
  Coverage   99.76%   99.76%           
=======================================
  Files          33       33           
  Lines        3410     3410           
  Branches      464      464           
=======================================
  Hits         3402     3402           
  Misses          5        5           
  Partials        3        3           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant