bd daemon sync: 2026-01-12 18:22:36

This commit is contained in:
beads/refinery 2026-01-12 18:22:36 -08:00 • committed by Steve Yegge
commit 6500a06937

View file

@ -1077,7 +1077,7 @@
{"id":"bd-dvd","title":"GetNextChildID doesn't attempt parent resurrection from JSONL history","description":"When creating a child issue with --parent flag, GetNextChildID fails immediately if parent doesn't exist in DB, without attempting to resurrect it from JSONL history. This breaks the intended resurrection workflow and causes 'parent issue X does not exist' errors even when the parent exists in JSONL.\n\nRelated to GH #334 and #278.\n\nCurrent behavior:\n- GetNextChildID checks if parent exists in DB\n- If not found, returns error immediately\n- No resurrection attempt\n\nExpected behavior:\n- GetNextChildID should call TryResurrectParent before failing\n- Parent should be restored as tombstone if found in JSONL history\n- Child creation should succeed if resurrection succeeds\n\nImpact: Users cannot create child issues for parents that were deleted but exist in JSONL history.","status":"closed","priority":1,"issue_type":"bug","created_at":"2025-11-21T10:02:51.496365-05:00","updated_at":"2025-11-22T14:57:44.534796097-05:00","closed_at":"2025-11-21T15:09:02.731171-05:00"}
{"id":"bd-dvw8","title":"GH#523: Fix closed issues missing closed_at timestamp during db upgrade","description":"Old beads databases may have closed issues without closed_at timestamps, causing 'closed issues must have closed_at timestamp' validation errors on upgrade to 0.29.0. Need defensive handling during import. See: https://github.com/steveyegge/beads/issues/523","status":"closed","priority":1,"issue_type":"bug","created_at":"2025-12-14T16:31:30.421555-08:00","updated_at":"2025-12-16T14:39:19.052482-08:00","closed_at":"2025-12-14T17:29:34.746247-08:00"}
{"id":"bd-dwh","title":"Implement or remove ExpectExit/ExpectStdout verification fields","description":"The Verification struct in internal/types/workflow.go has ExpectExit and ExpectStdout fields that are never used by workflowVerifyCmd. Either implement the functionality or remove the dead fields.","status":"tombstone","priority":3,"issue_type":"task","created_at":"2025-12-17T22:23:02.708627-08:00","updated_at":"2025-12-25T01:21:01.952723-08:00","deleted_at":"2025-12-25T01:21:01.952723-08:00","deleted_by":"batch delete","delete_reason":"batch delete","original_type":"task"}
{"id":"bd-dx2dc","title":"Refactor merge queue to use wisps instead of .beads/mq/ JSON files","description":"## The Heresy\n\nThe merge queue implementation in gastown uses a side-channel JSON storage (.beads/mq/*.json) instead of the beads system.\n\n## Architectural Intent (PRIMING.md)\n\n- MRs should be beads with type=merge-request\n- bd ready --type=merge-request is how Refinery should find work\n- Wisps are designed for 'high-volume, highly decomposed work streams' with 'no git overhead, fast writes'\n\n## Current Reality (gastown/internal/mrqueue/mrqueue.go)\n\n- Separate MR struct duplicating Issue fields\n- JSON files in .beads/mq/ directory\n- Comment says 'avoids sync overhead' but files get committed to git anyway\n- Cleanup doesn't work - stale entries accumulate (see PR #1035)\n\n## Evidence of Breakage\n\nPR #1035 shows 28 stale .beads/mq/*.json files from Dec 30 - all referencing branches that no longer exist. The 'delete after merge' cleanup failed silently.\n\n## Proposed Fix\n\n1. MRs become wisps with type=merge-request\n2. Refinery uses bd ready --type=merge-request\n3. Remove mrqueue package from gastown\n4. Wisps handle lifecycle (squash to digests, proper cleanup)\n\n## Affected Code\n\n- gastown/internal/mrqueue/ (delete)\n- gastown refinery (use bd commands)\n- gt done (create wisp instead of mq file)\n\n## Why This Matters\n\nSide-channel state violates 'discover, don't track' principle. The merge queue IS the validation/consensus layer per HOP architecture - it should be in beads, not beside it.","status":"open","priority":2,"issue_type":"task","owner":"steve.yegge@gmail.com","created_at":"2026-01-12T18:17:00.242528-08:00","created_by":"beads/crew/emma","updated_at":"2026-01-12T18:17:00.242528-08:00"}
{"id":"bd-dx2dc","title":"Refactor merge queue to use wisps instead of .beads/mq/ JSON files","description":"## The Heresy\n\nThe merge queue implementation in gastown uses a side-channel JSON storage (.beads/mq/*.json) instead of the beads system.\n\n## Architectural Intent (PRIMING.md)\n\n- MRs should be beads with type=merge-request\n- bd ready --type=merge-request is how Refinery should find work\n- Wisps are designed for 'high-volume, highly decomposed work streams' with 'no git overhead, fast writes'\n\n## Current Reality (gastown/internal/mrqueue/mrqueue.go)\n\n- Separate MR struct duplicating Issue fields\n- JSON files in .beads/mq/ directory\n- Comment says 'avoids sync overhead' but files get committed to git anyway\n- Cleanup doesn't work - stale entries accumulate (see PR #1035)\n\n## Evidence of Breakage\n\nPR #1035 shows 28 stale .beads/mq/*.json files from Dec 30 - all referencing branches that no longer exist. The 'delete after merge' cleanup failed silently.\n\n## Proposed Fix\n\n1. MRs become wisps with type=merge-request\n2. Refinery uses bd ready --type=merge-request\n3. Remove mrqueue package from gastown\n4. Wisps handle lifecycle (squash to digests, proper cleanup)\n\n## Affected Code\n\n- gastown/internal/mrqueue/ (delete)\n- gastown refinery (use bd commands)\n- gt done (create wisp instead of mq file)\n\n## Why This Matters\n\nSide-channel state violates 'discover, don't track' principle. The merge queue IS the validation/consensus layer per HOP architecture - it should be in beads, not beside it.","status":"in_progress","priority":2,"issue_type":"task","owner":"steve.yegge@gmail.com","created_at":"2026-01-12T18:17:00.242528-08:00","created_by":"beads/crew/emma","updated_at":"2026-01-12T18:22:36.033785-08:00"}
{"id":"bd-dx6e","title":"bd activity --town: Aggregated town feed","description":"Add --town flag to bd activity for aggregated cross-rig feed.\n\nBehavior:\n- Uses routes.jsonl to discover all rigs\n- Aggregates activity from all rig beads\n- Essential for Deacon's convoy tracking and cross-rig dep resolution\n\nUsage:\n bd activity --follow --town # Stream all rig activity\n\nRelated: hq-7h8jx (Convoy System epic in town beads)","status":"closed","priority":2,"issue_type":"task","created_at":"2025-12-29T18:47:03.057447-08:00","created_by":"mayor","updated_at":"2025-12-29T21:10:00.71961-08:00","closed_at":"2025-12-29T21:10:00.71961-08:00","close_reason":"Implemented --town flag for bd activity. Aggregates mutations from all rig daemons discovered via routes.jsonl."}
{"id":"bd-dx7m3","title":"Session ended: gt-beads-crew-emma","status":"closed","priority":2,"issue_type":"event","owner":"steve.yegge@gmail.com","created_at":"2026-01-12T02:29:21.865951-08:00","created_by":"beads/crew/emma","updated_at":"2026-01-12T02:29:21.931598-08:00","closed_at":"2026-01-12T02:29:21.931598-08:00","close_reason":"auto-closed session cost wisp","ephemeral":true}
{"id":"bd-dxdn","title":"bd ready taking 5 seconds with 132 issues (89 closed)","description":"User reports bd ready is annoyingly slow on M2 Mac - 5 seconds for 132 issues (89 closed). Started noticing after hash-based IDs update. Need to investigate performance regression. Reported in GH #243.","notes":"Root cause identified: Not a query performance issue, but stale daemon locks causing 5s timeout delays.\n\nFixed in bd-ndyz (closed) via 5 sub-issues:\n- bd-expt: Fast-fail socket checks (200ms timeout)\n- bd-wgu4: Lock probe before RPC attempts\n- bd-1mzt: Self-heal stale artifacts\n- bd-vcg5: Panic recovery + socket cleanup\n- bd-j7e2: RPC diagnostics (BD_RPC_DEBUG)\n\nAll fixes merged. Ready for v0.22.2 release.","status":"closed","priority":1,"issue_type":"bug","created_at":"2025-11-07T00:26:30.359512-08:00","updated_at":"2025-11-08T13:17:08.766029-08:00","closed_at":"2025-11-08T02:35:47.956638-08:00"}