feat(tasks): Consolidate schedule management into task_schedules

- Add schedule CRUD endpoints to /api/tasks/schedules
- Add Schedules section to TasksDashboard with edit/delete/bulk actions
- Deprecate job_schedules table (entries disabled in DB)
- Mark CrawlSchedulePage as deprecated (removed from menu)
- Add deprecation comments to legacy schedule methods in api.ts
- Add migration comments to workers.ts explaining consolidation

Key changes:
- Schedule management now at /admin/tasks instead of /admin/schedule
- task_schedules uses interval_hours (simpler than base_interval_minutes + jitter)
- All schedule routes placed before /:id to avoid Express route conflicts

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

backend/src/routes/tasks.ts

@@ -3,6 +3,24 @@
  *
  * Endpoints for managing worker tasks, viewing capacity metrics,
  * and generating batch tasks.
+ *
+ * SCHEDULE MANAGEMENT (added 2025-12-12):
+ * This file now contains the canonical schedule management endpoints.
+ * The job_schedules table has been deprecated and all schedule management
+ * is now consolidated into task_schedules:
+ *
+ * Schedule endpoints:
+ *   GET    /api/tasks/schedules           - List all schedules
+ *   POST   /api/tasks/schedules           - Create new schedule
+ *   GET    /api/tasks/schedules/:id       - Get schedule by ID
+ *   PUT    /api/tasks/schedules/:id       - Update schedule
+ *   DELETE /api/tasks/schedules/:id       - Delete schedule
+ *   DELETE /api/tasks/schedules           - Bulk delete schedules
+ *   POST   /api/tasks/schedules/:id/run-now - Trigger schedule immediately
+ *   POST   /api/tasks/schedules/:id/toggle  - Toggle schedule enabled/disabled
+ *
+ * Note: Schedule routes are defined BEFORE /:id to avoid route conflicts
+ * (Express matches routes in order, and "schedules" would match /:id otherwise)
  */
 
 import { Router, Request, Response } from 'express';
@@ -131,6 +149,366 @@ router.get('/capacity/:role', async (req: Request, res: Response) => {
   }
 });
 
+// ============================================================
+// SCHEDULE MANAGEMENT ROUTES
+// (Must be before /:id to avoid route conflicts)
+// ============================================================
+
+/**
+ * GET /api/tasks/schedules
+ * List all task schedules
+ */
+router.get('/schedules', async (req: Request, res: Response) => {
+  try {
+    const enabledOnly = req.query.enabled === 'true';
+
+    let query = `
+      SELECT id, name, role, description, enabled, interval_hours,
+             priority, state_code, platform, last_run_at, next_run_at,
+             last_task_count, last_error, created_at, updated_at
+      FROM task_schedules
+    `;
+
+    if (enabledOnly) {
+      query += ` WHERE enabled = true`;
+    }
+
+    query += ` ORDER BY name`;
+
+    const result = await pool.query(query);
+    res.json({ schedules: result.rows });
+  } catch (error: unknown) {
+    console.error('Error listing schedules:', error);
+    res.status(500).json({ error: 'Failed to list schedules' });
+  }
+});
+
+/**
+ * DELETE /api/tasks/schedules
+ * Bulk delete schedules
+ *
+ * Body:
+ * - ids: number[] (required) - array of schedule IDs to delete
+ * - all: boolean (optional) - if true, delete all schedules (ids ignored)
+ */
+router.delete('/schedules', async (req: Request, res: Response) => {
+  try {
+    const { ids, all } = req.body;
+
+    let result;
+
+    if (all === true) {
+      // Delete all schedules
+      result = await pool.query(`
+        DELETE FROM task_schedules RETURNING id, name
+      `);
+    } else if (Array.isArray(ids) && ids.length > 0) {
+      // Delete specific schedules by IDs
+      result = await pool.query(`
+        DELETE FROM task_schedules WHERE id = ANY($1) RETURNING id, name
+      `, [ids]);
+    } else {
+      return res.status(400).json({
+        error: 'Either provide ids array or set all=true',
+      });
+    }
+
+    res.json({
+      success: true,
+      deleted_count: result.rowCount,
+      deleted: result.rows,
+      message: `Deleted ${result.rowCount} schedule(s)`,
+    });
+  } catch (error: unknown) {
+    console.error('Error bulk deleting schedules:', error);
+    res.status(500).json({ error: 'Failed to delete schedules' });
+  }
+});
+
+/**
+ * POST /api/tasks/schedules
+ * Create a new schedule
+ *
+ * Body:
+ * - name: string (required, unique)
+ * - role: TaskRole (required)
+ * - description: string (optional)
+ * - enabled: boolean (default true)
+ * - interval_hours: number (required)
+ * - priority: number (default 0)
+ * - state_code: string (optional)
+ * - platform: string (optional)
+ */
+router.post('/schedules', async (req: Request, res: Response) => {
+  try {
+    const {
+      name,
+      role,
+      description,
+      enabled = true,
+      interval_hours,
+      priority = 0,
+      state_code,
+      platform,
+    } = req.body;
+
+    if (!name || !role || !interval_hours) {
+      return res.status(400).json({
+        error: 'name, role, and interval_hours are required',
+      });
+    }
+
+    // Calculate next_run_at based on interval
+    const nextRunAt = new Date(Date.now() + interval_hours * 60 * 60 * 1000);
+
+    const result = await pool.query(`
+      INSERT INTO task_schedules
+        (name, role, description, enabled, interval_hours, priority, state_code, platform, next_run_at)
+      VALUES ($1, $2, $3, $4, $5, $6, $7, $8, $9)
+      RETURNING id, name, role, description, enabled, interval_hours,
+                priority, state_code, platform, last_run_at, next_run_at,
+                last_task_count, last_error, created_at, updated_at
+    `, [name, role, description, enabled, interval_hours, priority, state_code, platform, nextRunAt]);
+
+    res.status(201).json(result.rows[0]);
+  } catch (error: any) {
+    if (error.code === '23505') {
+      // Unique constraint violation
+      return res.status(409).json({ error: 'A schedule with this name already exists' });
+    }
+    console.error('Error creating schedule:', error);
+    res.status(500).json({ error: 'Failed to create schedule' });
+  }
+});
+
+/**
+ * GET /api/tasks/schedules/:id
+ * Get a specific schedule by ID
+ */
+router.get('/schedules/:id', async (req: Request, res: Response) => {
+  try {
+    const scheduleId = parseInt(req.params.id, 10);
+
+    const result = await pool.query(`
+      SELECT id, name, role, description, enabled, interval_hours,
+             priority, state_code, platform, last_run_at, next_run_at,
+             last_task_count, last_error, created_at, updated_at
+      FROM task_schedules
+      WHERE id = $1
+    `, [scheduleId]);
+
+    if (result.rows.length === 0) {
+      return res.status(404).json({ error: 'Schedule not found' });
+    }
+
+    res.json(result.rows[0]);
+  } catch (error: unknown) {
+    console.error('Error getting schedule:', error);
+    res.status(500).json({ error: 'Failed to get schedule' });
+  }
+});
+
+/**
+ * PUT /api/tasks/schedules/:id
+ * Update an existing schedule
+ */
+router.put('/schedules/:id', async (req: Request, res: Response) => {
+  try {
+    const scheduleId = parseInt(req.params.id, 10);
+    const {
+      name,
+      role,
+      description,
+      enabled,
+      interval_hours,
+      priority,
+      state_code,
+      platform,
+    } = req.body;
+
+    // Build dynamic update query
+    const updates: string[] = [];
+    const values: any[] = [];
+    let paramIndex = 1;
+
+    if (name !== undefined) {
+      updates.push(`name = $${paramIndex++}`);
+      values.push(name);
+    }
+    if (role !== undefined) {
+      updates.push(`role = $${paramIndex++}`);
+      values.push(role);
+    }
+    if (description !== undefined) {
+      updates.push(`description = $${paramIndex++}`);
+      values.push(description);
+    }
+    if (enabled !== undefined) {
+      updates.push(`enabled = $${paramIndex++}`);
+      values.push(enabled);
+    }
+    if (interval_hours !== undefined) {
+      updates.push(`interval_hours = $${paramIndex++}`);
+      values.push(interval_hours);
+
+      // Recalculate next_run_at if interval changed
+      const nextRunAt = new Date(Date.now() + interval_hours * 60 * 60 * 1000);
+      updates.push(`next_run_at = $${paramIndex++}`);
+      values.push(nextRunAt);
+    }
+    if (priority !== undefined) {
+      updates.push(`priority = $${paramIndex++}`);
+      values.push(priority);
+    }
+    if (state_code !== undefined) {
+      updates.push(`state_code = $${paramIndex++}`);
+      values.push(state_code || null);
+    }
+    if (platform !== undefined) {
+      updates.push(`platform = $${paramIndex++}`);
+      values.push(platform || null);
+    }
+
+    if (updates.length === 0) {
+      return res.status(400).json({ error: 'No fields to update' });
+    }
+
+    updates.push('updated_at = NOW()');
+    values.push(scheduleId);
+
+    const result = await pool.query(`
+      UPDATE task_schedules
+      SET ${updates.join(', ')}
+      WHERE id = $${paramIndex}
+      RETURNING id, name, role, description, enabled, interval_hours,
+                priority, state_code, platform, last_run_at, next_run_at,
+                last_task_count, last_error, created_at, updated_at
+    `, values);
+
+    if (result.rows.length === 0) {
+      return res.status(404).json({ error: 'Schedule not found' });
+    }
+
+    res.json(result.rows[0]);
+  } catch (error: any) {
+    if (error.code === '23505') {
+      return res.status(409).json({ error: 'A schedule with this name already exists' });
+    }
+    console.error('Error updating schedule:', error);
+    res.status(500).json({ error: 'Failed to update schedule' });
+  }
+});
+
+/**
+ * DELETE /api/tasks/schedules/:id
+ * Delete a schedule
+ */
+router.delete('/schedules/:id', async (req: Request, res: Response) => {
+  try {
+    const scheduleId = parseInt(req.params.id, 10);
+
+    const result = await pool.query(`
+      DELETE FROM task_schedules WHERE id = $1 RETURNING id, name
+    `, [scheduleId]);
+
+    if (result.rows.length === 0) {
+      return res.status(404).json({ error: 'Schedule not found' });
+    }
+
+    res.json({
+      success: true,
+      message: `Schedule "${result.rows[0].name}" deleted`,
+    });
+  } catch (error: unknown) {
+    console.error('Error deleting schedule:', error);
+    res.status(500).json({ error: 'Failed to delete schedule' });
+  }
+});
+
+/**
+ * POST /api/tasks/schedules/:id/run-now
+ * Manually trigger a scheduled task to run immediately
+ */
+router.post('/schedules/:id/run-now', async (req: Request, res: Response) => {
+  try {
+    const scheduleId = parseInt(req.params.id, 10);
+
+    // Get the schedule
+    const scheduleResult = await pool.query(`
+      SELECT id, name, role, state_code, platform, priority
+      FROM task_schedules WHERE id = $1
+    `, [scheduleId]);
+
+    if (scheduleResult.rows.length === 0) {
+      return res.status(404).json({ error: 'Schedule not found' });
+    }
+
+    const schedule = scheduleResult.rows[0];
+
+    // Create a task based on the schedule
+    const task = await taskService.createTask({
+      role: schedule.role,
+      platform: schedule.platform,
+      priority: schedule.priority + 10, // Boost priority for manual runs
+    });
+
+    // Update last_run_at on the schedule
+    await pool.query(`
+      UPDATE task_schedules
+      SET last_run_at = NOW(),
+          next_run_at = NOW() + (interval_hours || ' hours')::interval,
+          updated_at = NOW()
+      WHERE id = $1
+    `, [scheduleId]);
+
+    res.json({
+      success: true,
+      message: `Schedule "${schedule.name}" triggered`,
+      task,
+    });
+  } catch (error: unknown) {
+    console.error('Error running schedule:', error);
+    res.status(500).json({ error: 'Failed to run schedule' });
+  }
+});
+
+/**
+ * POST /api/tasks/schedules/:id/toggle
+ * Toggle a schedule's enabled status
+ */
+router.post('/schedules/:id/toggle', async (req: Request, res: Response) => {
+  try {
+    const scheduleId = parseInt(req.params.id, 10);
+
+    const result = await pool.query(`
+      UPDATE task_schedules
+      SET enabled = NOT enabled,
+          updated_at = NOW()
+      WHERE id = $1
+      RETURNING id, name, enabled
+    `, [scheduleId]);
+
+    if (result.rows.length === 0) {
+      return res.status(404).json({ error: 'Schedule not found' });
+    }
+
+    res.json({
+      success: true,
+      schedule: result.rows[0],
+      message: result.rows[0].enabled
+        ? `Schedule "${result.rows[0].name}" enabled`
+        : `Schedule "${result.rows[0].name}" disabled`,
+    });
+  } catch (error: unknown) {
+    console.error('Error toggling schedule:', error);
+    res.status(500).json({ error: 'Failed to toggle schedule' });
+  }
+});
+
+// ============================================================
+// TASK-SPECIFIC ROUTES (with :id parameter)
+// ============================================================
+
 /**
  * GET /api/tasks/:id
  * Get a specific task by ID

backend/src/routes/workers.ts

@@ -4,10 +4,25 @@
  * Provider-agnostic worker management and job monitoring.
  * Replaces legacy /api/dutchie-az/admin/schedules and /api/dutchie-az/monitor/* routes.
  *
+ * DEPRECATION NOTE (2025-12-12):
+ * This file still queries job_schedules for backwards compatibility with
+ * the /api/workers endpoints that display worker status. However, the
+ * job_schedules table is DEPRECATED - all entries have been disabled.
+ *
+ * Schedule management has been consolidated into task_schedules:
+ * - Use /api/tasks/schedules for schedule CRUD operations
+ * - Use TasksDashboard.tsx (/admin/tasks) for schedule management UI
+ * - task_schedules uses interval_hours (simpler than base_interval_minutes + jitter)
+ *
+ * The /api/workers endpoints remain useful for:
+ * - Monitoring active workers and job status
+ * - K8s scaling controls
+ * - Job history and logs
+ *
  * Endpoints:
  *   GET /api/workers                    - List all workers/schedules
  *   GET /api/workers/active             - List currently active workers
- *   GET /api/workers/schedule           - Get all job schedules
+ *   GET /api/workers/schedule           - Get all job schedules (DEPRECATED - use /api/tasks/schedules)
  *   GET /api/workers/:workerName        - Get specific worker details
  *   GET /api/workers/:workerName/scope  - Get worker's scope (states, etc.)
  *   GET /api/workers/:workerName/stats  - Get worker statistics

backend/src/tasks/task-worker.ts

@@ -11,10 +11,17 @@
  *   - Workers report heartbeats to worker_registry
  *   - Workers are ROLE-AGNOSTIC by default (can handle any task type)
  *
- * Stealth & Anti-Detection:
- *   PROXIES ARE REQUIRED - workers will fail to start if no proxies available.
+ * Stealth & Anti-Detection (LAZY INITIALIZATION):
+ *   Workers start IMMEDIATELY without waiting for proxies.
+ *   Stealth systems (proxies, fingerprints, preflights) are initialized
+ *   on first task claim, not at worker startup.
  *
- *   On startup, workers initialize the CrawlRotator which provides:
+ *   This allows workers to:
+ *   - Register and send heartbeats immediately
+ *   - Wait in main loop without blocking on proxy availability
+ *   - Initialize proxies/preflights only when tasks are actually available
+ *
+ *   On first task claim attempt, workers initialize the CrawlRotator which provides:
  *   - Proxy rotation: Loads proxies from `proxies` table, ALL requests use proxy
  *   - User-Agent rotation: Cycles through realistic browser fingerprints
  *   - Fingerprint rotation: Changes browser profile on blocks
@@ -34,11 +41,16 @@
  *
  * Environment:
  *   WORKER_ROLE - Which task role to process (optional, null = any task)
- *   WORKER_ID - Optional custom worker ID (auto-generated if not provided)
- *   POD_NAME - Kubernetes pod name (optional)
+ *   POD_NAME - K8s StatefulSet pod name (PRIMARY - use this for persistent identity)
+ *   WORKER_ID - Custom worker ID (fallback if POD_NAME not set)
  *   POLL_INTERVAL_MS - How often to check for tasks (default: 5000)
  *   HEARTBEAT_INTERVAL_MS - How often to update heartbeat (default: 30000)
  *   API_BASE_URL - Backend API URL for registration (default: http://localhost:3010)
+ *
+ * Worker Identity:
+ *   Workers use POD_NAME as their worker_id for persistent identity across restarts.
+ *   In K8s StatefulSet, POD_NAME = "scraper-worker-0" through "scraper-worker-7".
+ *   This ensures workers re-register with the same ID instead of creating new entries.
  */
 
 import { Pool } from 'pg';
@@ -209,6 +221,16 @@ export class TaskWorker {
   private preflightCurlResult: CurlPreflightResult | null = null;
   private preflightHttpResult: PuppeteerPreflightResult | null = null;
 
+  // ==========================================================================
+  // LAZY INITIALIZATION FLAGS
+  // ==========================================================================
+  // Stealth/proxy initialization is deferred until first task claim.
+  // Workers register immediately and enter main loop without blocking.
+  // ==========================================================================
+  private stealthInitialized: boolean = false;
+  private preflightsCompleted: boolean = false;
+  private initializingPromise: Promise<void> | null = null;
+
   constructor(role: TaskRole | null = null, workerId?: string) {
     this.pool = getPool();
     this.role = role;
@@ -293,9 +315,9 @@ export class TaskWorker {
 
   /**
    * Initialize stealth systems (proxy rotation, fingerprints)
-   * Called once on worker startup before processing any tasks.
+   * Called LAZILY on first task claim attempt (NOT at worker startup).
    *
-   * IMPORTANT: Proxies are REQUIRED. Workers will wait until proxies are available.
+   * IMPORTANT: Proxies are REQUIRED to claim tasks. This method waits until proxies are available.
    * Workers listen for PostgreSQL NOTIFY 'proxy_added' to wake up immediately when proxies are added.
    */
   private async initializeStealth(): Promise<void> {
@@ -482,6 +504,51 @@ export class TaskWorker {
     }
   }
 
+  /**
+   * Lazy initialization of stealth systems.
+   * Called BEFORE claiming first task (not at worker startup).
+   * This allows workers to register and enter main loop immediately.
+   *
+   * Returns true if initialization succeeded, false otherwise.
+   */
+  private async ensureStealthInitialized(): Promise<boolean> {
+    // Already initialized
+    if (this.stealthInitialized && this.preflightsCompleted) {
+      return true;
+    }
+
+    // Already initializing (prevent concurrent init attempts)
+    if (this.initializingPromise) {
+      await this.initializingPromise;
+      return this.stealthInitialized && this.preflightsCompleted;
+    }
+
+    console.log(`[TaskWorker] ${this.friendlyName} lazy-initializing stealth systems (first task claim)...`);
+
+    this.initializingPromise = (async () => {
+      try {
+        // Initialize proxy/fingerprint rotation
+        await this.initializeStealth();
+        this.stealthInitialized = true;
+
+        // Run dual-transport preflights
+        await this.runDualPreflights();
+        this.preflightsCompleted = true;
+
+        const preflightMsg = `curl=${this.preflightCurlPassed ? '✓' : '✗'} http=${this.preflightHttpPassed ? '✓' : '✗'}`;
+        console.log(`[TaskWorker] ${this.friendlyName} stealth ready (${preflightMsg})`);
+      } catch (err: any) {
+        console.error(`[TaskWorker] ${this.friendlyName} stealth init failed: ${err.message}`);
+        this.stealthInitialized = false;
+        this.preflightsCompleted = false;
+      }
+    })();
+
+    await this.initializingPromise;
+    this.initializingPromise = null;
+    return this.stealthInitialized && this.preflightsCompleted;
+  }
+
   /**
    * Register worker with the registry (get friendly name)
    */
@@ -615,25 +682,22 @@ export class TaskWorker {
 
   /**
    * Start the worker loop
+   *
+   * Workers start IMMEDIATELY without blocking on proxy/preflight init.
+   * Stealth systems are lazy-initialized on first task claim.
+   * This allows workers to register and send heartbeats even when proxies aren't ready.
    */
   async start(): Promise<void> {
     this.isRunning = true;
 
-    // Initialize stealth systems (proxy rotation, fingerprints)
-    await this.initializeStealth();
-
-    // Register with the API to get a friendly name
+    // Register with the API to get a friendly name (non-blocking)
     await this.register();
 
-    // Run dual-transport preflights
-    await this.runDualPreflights();
-
-    // Start registry heartbeat
+    // Start registry heartbeat immediately
     this.startRegistryHeartbeat();
 
     const roleMsg = this.role ? `for role: ${this.role}` : '(role-agnostic - any task)';
-    const preflightMsg = `curl=${this.preflightCurlPassed ? '✓' : '✗'} http=${this.preflightHttpPassed ? '✓' : '✗'}`;
-    console.log(`[TaskWorker] ${this.friendlyName} starting ${roleMsg} (${preflightMsg}, max ${this.maxConcurrentTasks} concurrent tasks)`);
+    console.log(`[TaskWorker] ${this.friendlyName} starting ${roleMsg} (stealth=lazy, max ${this.maxConcurrentTasks} concurrent tasks)`);
 
     while (this.isRunning) {
       try {
@@ -687,6 +751,20 @@ export class TaskWorker {
 
     // Try to claim more tasks if we have capacity
     if (this.canAcceptMoreTasks()) {
+      // =================================================================
+      // LAZY INITIALIZATION - Initialize stealth on first task claim
+      // Workers start immediately and init proxies only when needed
+      // =================================================================
+      if (!this.stealthInitialized) {
+        const initSuccess = await this.ensureStealthInitialized();
+        if (!initSuccess) {
+          // Init failed - wait and retry next loop
+          console.log(`[TaskWorker] ${this.friendlyName} stealth init failed, waiting before retry...`);
+          await this.sleep(30000);
+          return;
+        }
+      }
+
       // Pass preflight capabilities to only claim compatible tasks
       const task = await taskService.claimTask(
         this.role,
@@ -922,7 +1000,10 @@ async function main(): Promise<void> {
     process.exit(1);
   }
 
-  const workerId = process.env.WORKER_ID;
+  // Use POD_NAME for persistent identity in K8s StatefulSet
+  // This ensures workers keep the same ID across restarts
+  // Falls back to WORKER_ID, then generates UUID if neither is set
+  const workerId = process.env.POD_NAME || process.env.WORKER_ID;
   // Pass null for role-agnostic, or the specific role
   const worker = new TaskWorker(role || null, workerId);