diff options
| author | soryu <soryu@soryu.co> | 2026-02-01 00:42:53 +0000 |
|---|---|---|
| committer | soryu <soryu@soryu.co> | 2026-02-01 00:42:53 +0000 |
| commit | 96ad3af6051af69e2e8b34b35e8b40926bdd13a1 (patch) | |
| tree | 2e2aedd39c66dedf7da301273306a0c77440ecf4 /makima/migrations/20260201000001_enhanced_supervisor_state.sql | |
| parent | bb14010db99b40792372bfcb4348cf4984f30b3f (diff) | |
| download | soryu-96ad3af6051af69e2e8b34b35e8b40926bdd13a1.tar.gz soryu-96ad3af6051af69e2e8b34b35e8b40926bdd13a1.zip | |
feat: Implement Phase 3 Tasks 3.3 and 3.4 - Supervisor State Persistence and Restoration
Task 3.3: Supervisor State Persistence
- Add migration 20260201000001_enhanced_supervisor_state.sql with new columns:
- state (supervisor state enum)
- current_activity (description)
- progress (0-100)
- error_message (for failed states)
- spawned_task_ids (tasks created by supervisor)
- pending_questions (questions awaiting user response)
- restoration_count, last_restored_at, restoration_source (restoration tracking)
- Update SupervisorState model with new fields
- Add PendingQuestion struct for tracking unanswered questions
- Add SupervisorRestorationContext for returning restoration info
- Add StateValidationResult and StateRecoveryAction for state validation
State persistence functions in repository.rs:
- update_supervisor_detailed_state() - Update state, activity, progress
- add_supervisor_spawned_task() - Track spawned tasks
- add_supervisor_pending_question() - Track pending questions
- remove_supervisor_pending_question() - Clear answered questions
- save_supervisor_state_full() - Full state save (UPSERT)
- mark_supervisor_restored() - Increment restoration count
- get_supervisors_with_pending_questions() - Find supervisors with pending questions
- get_supervisor_state_for_restoration() - Load state for restoration
- validate_spawned_tasks() - Validate task consistency
- update_supervisor_phase() - Update on phase change
- update_supervisor_heartbeat_state() - Lightweight heartbeat update
State save points:
- On task spawn (save_state_on_task_spawn)
- On question asked (save_state_on_question_asked)
- On question answered (clear_pending_question)
- On phase change (save_state_on_phase_change)
- On heartbeat (update_supervisor_heartbeat_state)
Task 3.4: Supervisor Restoration Protocol
- Add restoration detection when supervisor starts with existing state
- Implement validate_supervisor_state() for state consistency checks
- Implement restore_supervisor() with validation and context generation
- Add redeliver_pending_questions() for re-delivering questions after crash
- Add generate_restoration_context_message() for Claude context injection
- Update resume_supervisor endpoint to return RestorationInfo
- Re-deliver pending questions when supervisor resumes
Restoration flow:
1. Daemon restarts or task reassigned
2. Load supervisor state from supervisor_states
3. If NOT FOUND: Start fresh, log warning
4. If FOUND: Validate state consistency
5. If INVALID: Start from last checkpoint
6. If VALID: Restore conversation history
7. Check for pending questions - re-deliver to user
8. Check for waiting tasks - resume waiting state
9. Send restoration context to Claude
10. Resume execution from last state
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Diffstat (limited to 'makima/migrations/20260201000001_enhanced_supervisor_state.sql')
| -rw-r--r-- | makima/migrations/20260201000001_enhanced_supervisor_state.sql | 56 |
1 files changed, 56 insertions, 0 deletions
diff --git a/makima/migrations/20260201000001_enhanced_supervisor_state.sql b/makima/migrations/20260201000001_enhanced_supervisor_state.sql new file mode 100644 index 0000000..5411b73 --- /dev/null +++ b/makima/migrations/20260201000001_enhanced_supervisor_state.sql @@ -0,0 +1,56 @@ +-- Enhanced supervisor state persistence for restoration after crashes. +-- Adds additional fields to supervisor_states to track detailed state for recovery. + +-- Add state tracking field (matches SupervisorStateEnum: initializing, idle, working, +-- waiting_for_user, waiting_for_tasks, blocked, completed, failed, interrupted) +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS state VARCHAR(50) NOT NULL DEFAULT 'initializing'; + +-- Add current activity description for monitoring +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS current_activity TEXT; + +-- Add progress percentage (0-100) +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS progress INTEGER DEFAULT 0 + CHECK (progress >= 0 AND progress <= 100); + +-- Add error message for failed states +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS error_message TEXT; + +-- Add spawned task IDs (tasks this supervisor has created) +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS spawned_task_ids UUID[] DEFAULT ARRAY[]::UUID[]; + +-- Add pending questions (questions waiting for user response) +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS pending_questions JSONB DEFAULT '[]'; + +-- Add restoration metadata +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS restoration_count INTEGER DEFAULT 0; + +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS last_restored_at TIMESTAMPTZ; + +ALTER TABLE supervisor_states + ADD COLUMN IF NOT EXISTS restoration_source VARCHAR(50); + +-- Index for finding supervisors by state (useful for finding blocked/failed supervisors) +CREATE INDEX IF NOT EXISTS idx_supervisor_states_state ON supervisor_states(state); + +-- Index for finding supervisors with pending questions +CREATE INDEX IF NOT EXISTS idx_supervisor_states_pending_questions + ON supervisor_states USING gin(pending_questions) + WHERE pending_questions != '[]'::jsonb; + +COMMENT ON COLUMN supervisor_states.state IS 'Current supervisor state: initializing, idle, working, waiting_for_user, waiting_for_tasks, blocked, completed, failed, interrupted'; +COMMENT ON COLUMN supervisor_states.current_activity IS 'Human-readable description of current activity'; +COMMENT ON COLUMN supervisor_states.progress IS 'Progress percentage (0-100)'; +COMMENT ON COLUMN supervisor_states.error_message IS 'Error message when state is failed or blocked'; +COMMENT ON COLUMN supervisor_states.spawned_task_ids IS 'Array of task UUIDs spawned by this supervisor'; +COMMENT ON COLUMN supervisor_states.pending_questions IS 'Array of questions awaiting user response: [{id, question, choices, context, asked_at}]'; +COMMENT ON COLUMN supervisor_states.restoration_count IS 'Number of times this supervisor has been restored after interruption'; +COMMENT ON COLUMN supervisor_states.last_restored_at IS 'Timestamp of last restoration'; +COMMENT ON COLUMN supervisor_states.restoration_source IS 'Source of last restoration: daemon_restart, task_reassignment, manual'; |
