CRITICAL FIXES: 1. Stop button now resets database FIRST (before pkill) - Database cleanup happens even if coordinator crashed - Prevents stale 'running' chunks blocking restart - Uses Node.js sqlite library (not CLI - Docker compatible) 2. UI enhancement - 4-state display - ⚡ Processing (running > 0) - ⏳ Pending (pending > 0, running = 0) - ✅ Complete (all completed) - ⏸️ Idle (no work queued) [NEW] - Shows pending chunk count when present TECHNICAL DETAILS: - Replaced sqlite3 CLI calls with proper Node.js API - Fixed permissions: chown 1001:1001 cluster/ for container write - Database-first logic: reset → pkill → verify - Detailed logging for each operation step FILES CHANGED: - app/api/cluster/control/route.ts (database operations refactored) - app/cluster/page.tsx (4-state UI display) VERIFIED: - Stop button successfully reset 3 'running' chunks → 'pending' - UI correctly shows Idle state after Stop - Container logs show detailed operation flow - Database operations work in Docker environment DEPLOYMENT: - Container rebuilt with fixed code - Tested with real stale database (3 running chunks) - All operations working correctly
6.5 KiB
6.5 KiB