docs: record panel restart
This commit is contained in:
@@ -0,0 +1,58 @@
|
||||
# Task: Restart panel service after dashboard update
|
||||
|
||||
## Identity
|
||||
|
||||
- Task ID: 20260902-restart-panel-c7e921
|
||||
- Mode: Feature
|
||||
- Branch: main
|
||||
- Worktree: /Users/inmanx/Documents/lwltAPI
|
||||
- Base commit: 0d63201b29fe430813dc493883c01d2cd375df31
|
||||
- Owner: codex
|
||||
- Status: Ready for integration
|
||||
|
||||
## Scope
|
||||
|
||||
- Restart the existing local control-plane panel on `127.0.0.1:8786` from current local `main` commit `0d63201`.
|
||||
- Preserve the standard database and all task/ERP state.
|
||||
- Verify the listener, liveness, readiness, database schema, panel route, and leadership-dashboard copy after restart.
|
||||
|
||||
## Intent And Constraints
|
||||
|
||||
- The user explicitly authorized this service restart in the current turn.
|
||||
- Target only the verified 8786 listener in `/Users/inmanx/Documents/lwltAPI`; do not touch the separate 8876 preview or any ERP/browser process.
|
||||
- Use the repository's existing `pnpm run dev` startup path so migrations run before the control plane listens.
|
||||
- Stop only the PID that matches the expected command and working directory; fail closed if either identity check differs.
|
||||
- Do not read or output `.env`, mutate runtime tasks, access ERP, deploy remotely, or send data externally.
|
||||
|
||||
## Outcome
|
||||
|
||||
- Confirmed the pre-restart service was healthy and the exact listener was PID `70004`, running `control-plane/src/server.ts` from the main worktree under the bundled `pnpm run dev` process.
|
||||
- Sent `SIGTERM` to verified PID `70004`; it exited gracefully and released port 8786.
|
||||
- Started the existing `pnpm run dev` command from current local `main`; migrations completed and the new listener is PID `77059`.
|
||||
- The restarted process serves the leadership-dashboard update committed at `0d63201`.
|
||||
- No database rows, ERP state, product source, configuration, or external system was changed.
|
||||
|
||||
## Verification
|
||||
|
||||
- Before restart, `GET /health/live` returned healthy and `GET /health/ready` returned HTTP 200 with `database=true`, `schema=true`, and required migration `017_user_business_route_authorizations`.
|
||||
- After restart, PID `77059` listens on `127.0.0.1:8786`; old PID `70004` no longer exists.
|
||||
- After restart, `GET /health/live` returned `{"ok":true,"service":"ltjt-control-plane"}`.
|
||||
- After restart, `GET /health/ready` returned HTTP 200 with `database=true`, `schema=true`, and required migration `017_user_business_route_authorizations`.
|
||||
- `GET /operations-dashboard` returned HTTP 200 and included the current “平台运行全景/平台运行看板” presentation.
|
||||
- AgentBus remained disconnected before and after restart; this pre-existing state does not block panel liveness or database readiness.
|
||||
- `check_project_docs.py`: passed.
|
||||
- `check_doc_drift.py --task-id 20260902-restart-panel-c7e921`: passed.
|
||||
- `node --run check:repo`: 10/10 passed.
|
||||
- `node --run check`: passed.
|
||||
- `node --run test:control-plane`: 153/153 passed.
|
||||
- `node --run test:legacy`: 256/256 passed.
|
||||
- `node --run build`: passed.
|
||||
- `git diff --check`: passed.
|
||||
|
||||
## Follow-ups
|
||||
|
||||
- Refresh the signed-in browser page to load the restarted control-plane module and corrected historical dashboard type totals.
|
||||
|
||||
## Promotion Candidates
|
||||
|
||||
- None. This was an authorized operational restart with no product or architecture change.
|
||||
Reference in New Issue
Block a user