#102 - ci.yml full-suite comment is stale — #95 closed but full suite still runs continue-on-error and fails - Mark/osint-hub - cyguin git: Adlerian Hierarchical Self-Actualization Syllogism

Mark commented

2026-08-14 07:05:54 +00:00

Owner

Config cleanup from post-deploy probe of #95 (kanban card t_311f173d).

ci.yml contains the comment "Run full suite (non-gating until #95 is fixed)" and the full suite step runs with continue-on-error: true. #95 was closed 2026-08-13, but the full suite still fails on origin/main HEAD b81dc67: 13 security, 15 integration, ~13-14 concurrent, and performance failures (identical on pre-fix baseline d9bbe4d — pre-existing, not regressions).

Expected

Either re-gate the full suite now that #95 is closed, or update the stale comment to reflect the current policy (e.g. what the remaining failures are tracked by).

Config cleanup from post-deploy probe of #95 (kanban card t_311f173d). ci.yml contains the comment "Run full suite (non-gating until #95 is fixed)" and the full suite step runs with `continue-on-error: true`. #95 was closed 2026-08-13, but the full suite still fails on origin/main HEAD b81dc67: 13 security, 15 integration, ~13-14 concurrent, and performance failures (identical on pre-fix baseline d9bbe4d — pre-existing, not regressions). ## Expected Either re-gate the full suite now that #95 is closed, or update the stale comment to reflect the current policy (e.g. what the remaining failures are tracked by).

Mark added the

chore

blocked:preflight

labels

2026-08-14 07:05:54 +00:00

Mark referenced this issue from a commit

2026-08-20 07:39:13 +00:00

ci: update full-suite/security step comments to reference tracking issue #102

Mark referenced this issue from a pull request that will close it,

2026-08-20 07:40:58 +00:00

ci.yml: update stale full-suite comment (Option B, #102) #106

Mark referenced this issue from a commit

2026-09-10 13:19:24 +00:00

ci: update stale #95 comment to reference tracking issue #102 (#106)

Mark added

triage-approved

and removed

blocked:preflight

chore

labels

2026-09-15 01:59:39 +00:00

Mark commented

2026-09-15 01:59:39 +00:00

Author

Owner

Triaged: ci.yml full-suite comment stale (refers to closed #95). Small chore; routes to fixit lane.

Mark commented

2026-09-18 15:41:49 +00:00

Author

Owner

Resolved by upstream evolution rather than direct edit. The "Run full suite (non-gating until #95 is fixed)" comment and step have been removed entirely from main — .forgejo/workflows/ci.yml on HEAD d9dc0f8 is the 3-job lint gate (backend-lint, frontend-lint, all-checks-pass).

The trimming landed via 5184b67 fix(ci): trim workflow to a deterministic 3-job lint gate, which is the third option implicit in the issue: rather than re-gate the full suite (option 1) or update the stale comment (option 2), the team chose to drop the full-suite/security-test jobs altogether because the runner's single-job serial timeout was already killing the runs and the lint signal was the deterministic, network-light gate we can trust. The new Why a thin job set: comment at the top of ci.yml documents the policy explicitly.

No continue-on-error: true remains on any full-suite/integration/security/perf step on main — the only remaining continue-on-error: true flags are on the frontend-lint ESLint/Prettier steps (lines 77, 87), which are unrelated to the original #95 reference.

Closing as resolved.

Resolved by upstream evolution rather than direct edit. The "Run full suite (non-gating until #95 is fixed)" comment and step have been removed entirely from `main` — `.forgejo/workflows/ci.yml` on HEAD `d9dc0f8` is the 3-job lint gate (`backend-lint`, `frontend-lint`, `all-checks-pass`). The trimming landed via `5184b67 fix(ci): trim workflow to a deterministic 3-job lint gate`, which is the third option implicit in the issue: rather than re-gate the full suite (option 1) or update the stale comment (option 2), the team chose to drop the full-suite/security-test jobs altogether because the runner's single-job serial timeout was already killing the runs and the lint signal was the deterministic, network-light gate we can trust. The new `Why a thin job set:` comment at the top of ci.yml documents the policy explicitly. No `continue-on-error: true` remains on any full-suite/integration/security/perf step on `main` — the only remaining `continue-on-error: true` flags are on the frontend-lint ESLint/Prettier steps (lines 77, 87), which are unrelated to the original #95 reference. Closing as resolved.

Mark closed this issue

2026-09-18 15:41:49 +00:00

Rows
Columns

ci.yml full-suite comment is stale — #95 closed but full suite still runs continue-on-error and fails #102

Expected