ci: run browser journeys on three shards per engine - #280
Merged
Merged
Conversation
loganj
force-pushed
the
larry/webkit-three-shards
branch
from
September 25, 2026 18:39
c244dd2 to
63a9059
Compare
loganj
force-pushed
the
larry/webkit-three-shards
branch
from
September 25, 2026 18:57
63a9059 to
11a903b
Compare
loganj
marked this pull request as ready for review
September 25, 2026 19:10
With two shards per engine, a WebKit shard ran about 12.5 minutes of tests and a Chromium shard up to about 11. With 1-2.5 minutes of runner setup, jobs often reached the 15-minute limit and were cancelled while their tests were still passing. Keep the limit, so a hang still fails fast, and use three shards per engine. Signed-off-by: Larry <627498bd4bd1f281a16431e3c6cce3b5c25b6692798c78672298aefbf2f8f8b5@buzz.block.builderlab.xyz>
loganj
force-pushed
the
larry/webkit-three-shards
branch
from
September 25, 2026 19:11
11a903b to
d1097a5
Compare
kalvinnchau
approved these changes
Sep 25, 2026
zrmarley
added a commit
that referenced
this pull request
Sep 25, 2026
…-image * origin/main: (23 commits) fix(agents): recover status polling and scope failure diagnostics (#283) Share avatar editing across community profiles and managed agents (#271) feat(profiles): archive, unarchive and delete agents from the profile pane (#256) ci: run browser journeys on three shards per engine (#280) ci: publish scheduled macOS test prereleases (#262) feat: add private text feedback plugin (#242) 🤖 docs: add pre-PR checklist and review guidance to AGENTS.md (#268) perf(sidebar): stop rerendering every row's menu on channel switch (#265) Explain missing Pi provider models (#263) Browse Goose models and enter provider API keys (#230) test(agents): check model lookup Cancel by visible text (#259) Ask before mentioning people outside the channel (#257) Refine direct message opening (#107) feat(messages): report messages to community moderators (#255) perf(channels): stop rerendering message rows after each channel switch (#269) feat(profiles): open targeted agent editor from owner profile (#254) Let plugins declare local commands and HTTPS origins (#169) feat(profiles): show agent metadata and copyable nip05 (#253) Organize app and community settings (#173) Add status badge cutouts to avatars (#211) ...
johnmatthewtennant
pushed a commit
that referenced
this pull request
Sep 28, 2026
* origin/main: feat: attach sanitized image and opt-in diagnostics to feedback (#245) test: repair three baseline Vitest failures (#276) fix(agents): recover status polling and scope failure diagnostics (#283) Share avatar editing across community profiles and managed agents (#271) feat(profiles): archive, unarchive and delete agents from the profile pane (#256) ci: run browser journeys on three shards per engine (#280) Signed-off-by: Sol <49aa1f65411fd096d2e2ec144f1e7aa36fdc76d1b907cfdf7be000c66f9d3b8e@buzz.block.builderlab.xyz> # Conflicts: # src/bundled/agents/AgentEditor.tsx
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
2 participants
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.
🤖
Summary
CI requiredfails and the PR shows red for no real reason. In recent runs this happened on main, on fix(status): reopen a Today status as Today near 16:00 #275, and on at least one other PR branch.Details
With two runners per engine, a WebKit runner spent about 12.5 minutes on tests and runner setup took another 50–140 seconds. Chromium is faster but not safe either: across its last 34 successful jobs, the median was 11.4 minutes and the slowest was 13.45 minutes.
Estimated test time per runner uses per-test times from a recent passing run (test: repair three baseline Vitest failures #276). Each runner has two workers, so run time is about half of the total test-minutes:
Playwright gives each runner whole test files, so the split is uneven. Four runners would save only about 3 more test-minutes on the slowest WebKit runner, so this PR uses three.
This adds two runners per CI run. Job names change from
(webkit, 1/2)to(webkit, 1/3). The only required check isCI required, which still requires every shard to pass.tests/integration/browser-ci.test.mjsnow expects shards[1, 2, 3]. Its coverage test runs each shard's actual command and still checks that every functional test runs exactly once.docs/contributing.mdanddocs/browser-testing.mdnow describe three shards per engine.