CI Health Dashboard

[igor-kupczynski]

CI Health Dashboard

Window: last 14 days (trend + pass rate) · tables: last 24h · updated 2026-09-04T07:12:04Z · auto-generated, do not edit by hand.

Gating-CI pass rate — PR: 85% (2309/2730) · main: 83% (138/166)

Gating-CI pass-rate trend

xychart-beta
  title "Gating-CI pass rate (%) per day"
  x-axis [21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 1, 2, 3, 4]
  y-axis "pass rate %" 0 --> 100
  line "CI" [82, 73, 93, 88, 93, 88, 92, 87, 100, 71, 86, 78, 88, 76, 68]
  line "main" [76, 100, 100, 81, 100, 88, 100, 62, 62, 62, 90, 91, 71, 76, 100]

Loading

X-axis = day of month (Aug 21 → Sep 04). Two lines: CI (PR gating-CI runs, generally the upper line) and main (post-merge main runs, lower). Y-axis = % of that day's gating-CI runs that passed.

Top 10 failing jobs (last 24h)

#	job	workflow	fails	recovered	runs	fail rate	flaky?	scope	cause
1	build	frontend / docs	9	0	24	38%	flaky	PR	infra/CI — frontend/docs prettier drift on migration-guide-python-v2.mdx
2	lint	ruby	8	0	28	29%	flaky	PR	infra/CI — Ruby protobuf/REST generated bindings out of date
3	unit	test	7	0	38	18%	flaky	main + PR	product bug — duplicate dagoperator methods block test/unit compile
4	test-templates	cli-e2e-tests	6	0	14	43%	flaky	main + PR	flaky test — CLI TestQuickstartTemplates parent E2E template timing
5	integration	test	6	0	38	16%	flaky	main + PR	product bug — duplicate dagoperator methods block test/integration generate
6	generate	test	6	0	38	16%	flaky	PR	product bug — test/generate fails golangci-lint on duplicate dagoperator methods
7	test	ruby	5	0	28	18%	flaky	main + PR	product bug — duplicate method reportBlockedOnDurableEvents in dagoperator/dag.go
8	cypress	frontend / app	5	0	33	15%	flaky	PR	product bug — duplicate dagoperator methods block ruby test job setup compile
9	test	python	5	0	44	11%	flaky	PR	product bug — Python durable sleep cancel replay KeyError on runtime payload
10	lint	lint all	5	0	46	11%	flaky	PR	infra/CI — lint all pre-commit failure (changelog formatting drift)

Top 10 failing tests (last 24h)

#	test	job	fails	runs	fail rate	flaky?	scope	cause
1	examples/conditions/test_conditions.py::test_waits	test	10	44	23%	flaky	PR	flaky test — conditions test_waits times out waiting for workflow start (60s)
2	(unparsed)	build	9	24	38%	flaky	PR	infra/CI — frontend/docs prettier drift on migration-guide-python-v2.mdx
3	examples/bug_tests/payload_bug_on_replay/test_payload_replay_bug.py::test_payload_replay_bug	test	9	44	20%	flaky	main + PR	product bug — Python payload replay bug KeyError step2 on replay
4	(unparsed)	lint	7	28	25%	flaky	PR	infra/CI — Ruby protobuf/REST generated bindings out of date
5	(unparsed)	lint	7	33	21%	flaky	PR	infra/CI — TypeScript protobuf/REST generated bindings out of date
6	(unparsed)	lint	7	44	16%	flaky	PR	infra/CI — Python protobuf/REST generated bindings out of date
7	examples/durable/test_durable.py::test_durable_sleep_cancel_replay	test	7	44	16%	flaky	main + PR	product bug — Python durable sleep cancel replay KeyError on runtime payload
8	TestQuickstartTemplates	test-templates	5	14	36%	flaky	main + PR	flaky test — CLI TestQuickstartTemplates parent E2E template timing
9	TestQuickstartTemplates/simple_go_go	test-templates	5	14	36%	flaky	main + PR	flaky test — CLI TestQuickstartTemplates/simple_go_go E2E template timing
10	(unparsed)	lint	5	46	11%	flaky	PR	infra/CI — lint all pre-commit failure (changelog formatting drift)

Recent CI-health wins (ci-health)

Recently merged

• test(e2e): wait for durable terminal statuses #4239
• test(sdk/python): stabilize bulk replay polling #4238
• chore: pin rabbit version #4218
• test(e2e): wait for durable child failure status #4213
• fix(olap): don't lose status updates when replaying completed tasks #4165

Open

No open ci-health PRs yet.

---

Trend and pass-rate totals cover the last 14 days; job/test tables cover the last 24h. fails = gating runs where the job/test failed · recovered = failed on a first attempt but passed on re-run (a flakiness signal) · runs = total gating runs of that workflow · fail rate = fails ÷ runs · flaky = recovered on re-run or intermittent across runs; deterministic = fails every time it runs · scope = whether failures were seen on PR, main, or main + PR.