Skip to content

main CI failure: pm-aggregator Typecheck step — 27 pre-existing type errors unmasked after #4827 fix #4848

Closed

Description

@dezren39

main CI failure — pm-aggregator Typecheck step (unmasked 2026-11-03)

What happened

The two failing aggregator tests tracked by issue #4827 were fixed and merged (PR #4842). On the post-merge main run 41616879803 the Run aggregator tests step passes — and the Typecheck step, which was skipped on every run where the test step failed first, now runs and fails with 27 pre-existing type errors across 12 files. These errors were invisible until the test step went green; none of them are in the files PR #4842 touched.

Errors (from gh run view 41616879803 --log-failed)

  • src/dns.ts(191,13): TS2531 Object is possibly 'null'
  • src/projects.ts(151,27): TS2345 number | null | undefined not assignable to number | undefined
  • src/push.ts(9,21): TS7016 no declaration file for module 'web-push'
  • test/charges-collect.test.ts: 6 errors — fixture rows missing method/source from ChargeRow (lines 72, 83, 93, 99)
  • test/clinepass-adapter.test.ts(207,27): TS2769 '"***"' not assignable to null
  • test/feed.test.ts: 6 errors — fixtures missing SessionRecord properties (701, 703, 712) and source: string not assignable to ChargeSourceKind (989, 999, 1080)
  • test/github-billing.test.ts(161,33): TS2365 operator > on number | { low, high }
  • test/overview.test.ts(20,3): TS2741 missing checkout in feed-health fixture
  • test/projects.test.ts: 4 errors — fetch stub missing preconnect (101, 112, 124), null not assignable to number (167)
  • test/quota-current.test.ts(613,78/82): TS18048 'a'/'b' possibly 'undefined'
  • test/registry-view.test.ts(62,14),(63,14): TS2554 expected 1 argument, got 0
  • test/token-adapter.test.ts(42,58): TS2769 fixture union not assignable to CredentialRef

Acceptance

  • A main-branch run of pm-aggregator tests completes with conclusion success (both the test step and the Typecheck step green).
  • Fixes should address the type errors at their source (guards/fixtures/declarations), not by weakening tsc config or excluding files.

Filed by repo queue worker lane A (task 6944 follow-up) after fixing issue #4827.

Activity

  1. added
    kind/bugSomething isn't working (prefixed form of the classic bug label)
    priority/highImportant and time-sensitive
    source/ci-watchFiled by the main-ci-failure-fix-trigger (fleet CI watch)
    queue/queuedHas a queued task in the Supabase agent queue
    task/small-fixQueue packet kind: bounded fix with reproduction and targeted tests
    work/assignedQueue worker lifecycle state
    queue/claimedQueue worker lifecycle state
    and removed
    queue/queuedHas a queued task in the Supabase agent queue
    on Nov 3, 2026
  2. dezren39 commented on Nov 3, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · picked up (OpenChamber scheduled lane)

  3. added
    queue/parkedQueue task is parked (waiting on CI, time, or an event)
    status/parkedDeliberately paused with a recorded resume condition
    and removed on Nov 3, 2026
  4. 9 remaining items

  5. dezren39 commented on Nov 3, 2026

    @dezren39
    MemberAuthor

    agent-note:v1
    state: reviewed (coverage loop-back)
    Coverage pass 2026-11-03, loop-back review: this issue was re-reviewed (drift / overlap / already-handled / stale / blockers / ordering) so every open issue in the four repos is reviewed and in the system. Verdict: still live; in the system as task 7170 (small_fix, claimed, priority 40) in repo-health-code-pilot. Labels audited and synced. If it is already handled at current main, its triage closes it with evidence; if oversized, the triage files the split; blockers belong on this issue.
    Found via: coverage loop-back 2026-11-03

  6. dezren39 commented on Nov 3, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · handed off

    • Agent: queue worker (repo-worker-code), lane D
    • Session / ID: task 7170
    • Found via: queue claim
    • Task: The fix for issue main CI failure: pm-aggregator Typecheck step — 27 pre-existing type errors unmasked after #4827 fix #4848 (pm-aggregator Typecheck: 27 pre-existing type errors) is assigned to the opencode execution lane (assignment re-confirmed this claim, 2026-11-03T14:49Z); checkpointed handed_off — the laptop claimant launches the bound opencode session, which drives task 7170 to its terminal state.
    • Expectation: the claimant session resumes from the recorded reproduction (27 errors across 12 files reproduced with bunx tsc --noEmit; worktree /home/user/git/fleet/7170, planned branch queue/7170-4848-aggregator-typecheck, no source edits yet) and fixes the errors at their source without weakening tsconfig.
  7. dezren39 commented on Nov 3, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · picked up (resume)

    • Agent: repo-worker-code, OpenChamber scheduled lane
    • Session / ID: queue task 7170 (attempt 4), lane 262eb403-3347-4a7c-ba6b-e7bc263ca853
    • Found via: atomic claim; resumed from task checkpoint handed_off
    • Revalidated: main CI failure: pm-aggregator Typecheck step — 27 pre-existing type errors unmasked after #4827 fix #4848 OPEN, no hard-stop labels; no competing open PR. Existing worktree ~/git/fleet/7170, branch queue/7170-4848-aggregator-typecheck had a clean tree and was based on 08c6e3bb4.
    • Fresh baseline: fast-forwarded worktree with current origin/main 1e8838fd3; just 3 of the original 27 errors remained, because intervening changes had fixed the rest.
    • Next: finish the remaining type-only fixes, verify, and open/complete the PR.
  8. added 2 commits that reference this issue on Nov 3, 2026
  9. removed
    queue/queuedHas a queued task in the Supabase agent queue
    queue/claimedQueue worker lifecycle state
    work/assignedQueue worker lifecycle state
    on Nov 3, 2026
  10. dezren39 commented on Nov 3, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · done (fixed, merged)

    • Agent: repo-worker-code, OpenChamber scheduled lane
    • Session / ID: queue task 7170 (attempt 4), lane 262eb403-3347-4a7c-ba6b-e7bc263ca853
    • Found via: atomic claim, resumed from checkpoint handed_off (worktree ~/git/fleet/7170 already existed and reproduced 27 errors at base 08c6e3bb4)
    • What landed: PR fix(aggregator): clear the last Typecheck errors on main (#4848) #5029 MERGED as 42ca3b3c0f7020b5ed631b7935247872c4b934ad. Only 3 of the 27 listed errors still reproduced after merging up to origin/main 1e8838fd3; intervening PRs had fixed the other 24 (src/dns.ts, src/projects.ts, src/push.ts and the test fixtures). Remaining fixes: rfa added to the HealthView fixture in test/overview.test.ts, and parsed?.sanitized?.dropped* optional chains in test/quota-current.test.ts (2 assertions). No tsconfig weakening, no exclusions, no behaviour change.
    • Verified locally (laptop): bunx tsc --noEmit clean; bun test test/overview.test.ts test/quota-current.test.ts 55 pass / 0 fail; full bun test 1365 pass / 0 fail across 98 files.
    • CI: this workflow has no pull_request trigger (on: push: branches:[main]), so no PR-tier signal exists for this path; the packet's acceptance is the post-merge main run. Post-merge run for the merge commit is 41677573258 (https://github.com/developing-today/code/actions/runs/41677573258) and was queued — the repo's Actions queue is backing up (older main runs still queued since 15:49Z). That run is the remaining verification, tracked by this issue's own auto-close/CI watch; no further code work is outstanding here.
    • Bookkeeping: sandbox backlog/sweep.log deferred to fleet reconciliation (OpenChamber scheduled lane). Issue auto-closed via Closes #4848; in-flight labels cleared.
  11. dezren39 commented on Nov 3, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · handed off

    • Agent: fleet monitor
    • Session / ID: queue task 7170 · repo-fleet-monitor run 2026-11-03 10:12 CST (note posted on a delayed retry — the original posting attempts were rejected by a GitHub secondary rate limit; this text is the corrected, re-verified record)
    • Found via: monitor reconciliation
    • Task: Claim/label divergence repair (Duty 1 addendum): at this run's verification, task 7170 was queued and unclaimed (no live lease) while this issue still carried queue/claimed + work/assigned — the claim had evaporated. The monitor reset the labels to the truth (queue/queued + status/needs-triage; since re-verified as landed). Last checkpoint stage on the task: handed_off. The task was also dead-in-queue (attempts at cap); the monitor doubled max_attempts 3 → 6 in the same run. Current task state at this posting: done.
    • Expectation: effort s — the work was not lost, only unheld; the task's subsequent state above is the proof it re-entered the system.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

kind/bugSomething isn't working (prefixed form of the classic bug label)priority/highImportant and time-sensitivequeue/queuedHas a queued task in the Supabase agent queuesource/ci-watchFiled by the main-ci-failure-fix-trigger (fleet CI watch)status/needs-triageNot yet triaged; needs a first looktask/small-fixQueue packet kind: bounded fix with reproduction and targeted tests

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions