Merge pull request #781 from gsxdsm/fix/improve-restart-recovery

feat: Add feature state reconciliation on server startup

apps/server/src/index.ts

@@ -368,24 +368,61 @@ eventHookService.initialize(events, settingsService, eventHistoryService, featur
     logger.warn('Failed to check for legacy settings migration:', err);
   }
 
-  // Apply logging settings from saved settings
+  // Fetch global settings once and reuse for logging config and feature reconciliation
+  let globalSettings: Awaited<ReturnType<typeof settingsService.getGlobalSettings>> | null = null;
   try {
-    const settings = await settingsService.getGlobalSettings();
+    globalSettings = await settingsService.getGlobalSettings();
-    if (settings.serverLogLevel && LOG_LEVEL_MAP[settings.serverLogLevel] !== undefined) {
-      setLogLevel(LOG_LEVEL_MAP[settings.serverLogLevel]);
-      logger.info(`Server log level set to: ${settings.serverLogLevel}`);
-    }
-    // Apply request logging setting (default true if not set)
-    const enableRequestLog = settings.enableRequestLogging ?? true;
-    setRequestLoggingEnabled(enableRequestLog);
-    logger.info(`HTTP request logging: ${enableRequestLog ? 'enabled' : 'disabled'}`);
   } catch (err) {
-    logger.warn('Failed to load logging settings, using defaults');
+    logger.warn('Failed to load global settings, using defaults');
+  }
+
+  // Apply logging settings from saved settings
+  if (globalSettings) {
+    try {
+      if (
+        globalSettings.serverLogLevel &&
+        LOG_LEVEL_MAP[globalSettings.serverLogLevel] !== undefined
+      ) {
+        setLogLevel(LOG_LEVEL_MAP[globalSettings.serverLogLevel]);
+        logger.info(`Server log level set to: ${globalSettings.serverLogLevel}`);
+      }
+      // Apply request logging setting (default true if not set)
+      const enableRequestLog = globalSettings.enableRequestLogging ?? true;
+      setRequestLoggingEnabled(enableRequestLog);
+      logger.info(`HTTP request logging: ${enableRequestLog ? 'enabled' : 'disabled'}`);
+    } catch (err) {
+      logger.warn('Failed to apply logging settings, using defaults');
+    }
   }
 
   await agentService.initialize();
   logger.info('Agent service initialized');
 
+  // Reconcile feature states on startup
+  // After any type of restart (clean, forced, crash), features may be stuck in
+  // transient states (in_progress, interrupted, pipeline_*) that don't match reality.
+  // Reconcile them back to resting states before the UI is served.
+  if (globalSettings) {
+    try {
+      if (globalSettings.projects && globalSettings.projects.length > 0) {
+        let totalReconciled = 0;
+        for (const project of globalSettings.projects) {
+          const count = await autoModeService.reconcileFeatureStates(project.path);
+          totalReconciled += count;
+        }
+        if (totalReconciled > 0) {
+          logger.info(
+            `[STARTUP] Reconciled ${totalReconciled} feature(s) across ${globalSettings.projects.length} project(s)`
+          );
+        } else {
+          logger.info('[STARTUP] Feature state reconciliation complete - no stale states found');
+        }
+      }
+    } catch (err) {
+      logger.warn('[STARTUP] Failed to reconcile feature states:', err);
+    }
+  }
+
   // Bootstrap Codex model cache in background (don't block server startup)
   void codexModelCacheService.getModels().catch((err) => {
     logger.error('Failed to bootstrap Codex model cache:', err);

apps/server/src/routes/auto-mode/index.ts

@@ -21,6 +21,7 @@ import { createFollowUpFeatureHandler } from './routes/follow-up-feature.js';
 import { createCommitFeatureHandler } from './routes/commit-feature.js';
 import { createApprovePlanHandler } from './routes/approve-plan.js';
 import { createResumeInterruptedHandler } from './routes/resume-interrupted.js';
+import { createReconcileHandler } from './routes/reconcile.js';
 
 /**
  * Create auto-mode routes.
@@ -81,6 +82,11 @@ export function createAutoModeRoutes(autoModeService: AutoModeServiceCompat): Ro
     validatePathParams('projectPath'),
     createResumeInterruptedHandler(autoModeService)
   );
+  router.post(
+    '/reconcile',
+    validatePathParams('projectPath'),
+    createReconcileHandler(autoModeService)
+  );
 
   return router;
 }

apps/server/src/routes/auto-mode/routes/reconcile.ts

@@ -0,0 +1,53 @@
+/**
+ * Reconcile Feature States Handler
+ *
+ * On-demand endpoint to reconcile all feature states for a project.
+ * Resets features stuck in transient states (in_progress, interrupted, pipeline_*)
+ * back to resting states (ready/backlog) and emits events to update the UI.
+ *
+ * This is useful when:
+ * - The UI reconnects after a server restart
+ * - A client detects stale feature states
+ * - An admin wants to force-reset stuck features
+ */
+
+import type { Request, Response } from 'express';
+import { createLogger } from '@automaker/utils';
+import type { AutoModeServiceCompat } from '../../../services/auto-mode/index.js';
+
+const logger = createLogger('ReconcileFeatures');
+
+interface ReconcileRequest {
+  projectPath: string;
+}
+
+export function createReconcileHandler(autoModeService: AutoModeServiceCompat) {
+  return async (req: Request, res: Response): Promise<void> => {
+    const { projectPath } = req.body as ReconcileRequest;
+
+    if (!projectPath) {
+      res.status(400).json({ error: 'Project path is required' });
+      return;
+    }
+
+    logger.info(`Reconciling feature states for ${projectPath}`);
+
+    try {
+      const reconciledCount = await autoModeService.reconcileFeatureStates(projectPath);
+
+      res.json({
+        success: true,
+        reconciledCount,
+        message:
+          reconciledCount > 0
+            ? `Reconciled ${reconciledCount} feature(s)`
+            : 'No features needed reconciliation',
+      });
+    } catch (error) {
+      logger.error('Error reconciling feature states:', error);
+      res.status(500).json({
+        error: error instanceof Error ? error.message : 'Unknown error',
+      });
+    }
+  };
+}

apps/server/src/services/auto-mode/compat.ts

@@ -88,6 +88,10 @@ export class AutoModeServiceCompat {
     return this.globalService.markAllRunningFeaturesInterrupted(reason);
   }
 
+  async reconcileFeatureStates(projectPath: string): Promise<number> {
+    return this.globalService.reconcileFeatureStates(projectPath);
+  }
+
   // ===========================================================================
   // PER-PROJECT OPERATIONS (delegated to facades)
   // ===========================================================================

apps/server/src/services/auto-mode/global-service.ts

@@ -205,4 +205,21 @@ export class GlobalAutoModeService {
       );
     }
   }
+
+  /**
+   * Reconcile all feature states for a project on server startup.
+   *
+   * Resets features stuck in transient states (in_progress, interrupted, pipeline_*)
+   * back to a resting state and emits events so the UI reflects corrected states.
+   *
+   * This should be called during server initialization to handle:
+   * - Clean shutdown: features already marked as interrupted
+   * - Forced kill / crash: features left in in_progress or pipeline_* states
+   *
+   * @param projectPath - The project path to reconcile
+   * @returns The number of features that were reconciled
+   */
+  async reconcileFeatureStates(projectPath: string): Promise<number> {
+    return this.featureStateManager.reconcileAllFeatureStates(projectPath);
+  }
 }

apps/server/src/services/feature-state-manager.ts

@@ -25,6 +25,7 @@ import {
 import { getFeatureDir, getFeaturesDir } from '@automaker/platform';
 import * as secureFs from '../lib/secure-fs.js';
 import type { EventEmitter } from '../lib/events.js';
+import type { AutoModeEventType } from './typed-event-bus.js';
 import { getNotificationService } from './notification-service.js';
 import { FeatureLoader } from './feature-loader.js';
 
@@ -268,20 +269,39 @@ export class FeatureStateManager {
   }
 
   /**
-   * Reset features that were stuck in transient states due to server crash.
+   * Shared helper that scans features in a project directory and resets any stuck
-   * Called when auto mode is enabled to clean up from previous session.
+   * in transient states (in_progress, interrupted, pipeline_*) back to resting states.
    *
-   * Resets:
+   * Also resets:
-   * - in_progress features back to ready (if has plan) or backlog (if no plan)
    * - generating planSpec status back to pending
    * - in_progress tasks back to pending
    *
-   * @param projectPath - The project path to reset features for
+   * @param projectPath - The project path to scan
+   * @param callerLabel - Label for log messages (e.g., 'resetStuckFeatures', 'reconcileAllFeatureStates')
+   * @returns Object with reconciledFeatures (id + status info), reconciledCount, and scanned count
    */
-  async resetStuckFeatures(projectPath: string): Promise<void> {
+  private async scanAndResetFeatures(
+    projectPath: string,
+    callerLabel: string
+  ): Promise<{
+    reconciledFeatures: Array<{
+      id: string;
+      previousStatus: string | undefined;
+      newStatus: string | undefined;
+    }>;
+    reconciledFeatureIds: string[];
+    reconciledCount: number;
+    scanned: number;
+  }> {
     const featuresDir = getFeaturesDir(projectPath);
-    let featuresScanned = 0;
+    let scanned = 0;
-    let featuresReset = 0;
+    let reconciledCount = 0;
+    const reconciledFeatureIds: string[] = [];
+    const reconciledFeatures: Array<{
+      id: string;
+      previousStatus: string | undefined;
+      newStatus: string | undefined;
+    }> = [];
 
     try {
       const entries = await secureFs.readdir(featuresDir, { withFileTypes: true });
@@ -289,7 +309,7 @@ export class FeatureStateManager {
       for (const entry of entries) {
         if (!entry.isDirectory()) continue;
 
-        featuresScanned++;
+        scanned++;
         const featurePath = path.join(featuresDir, entry.name, 'feature.json');
         const result = await readJsonWithRecovery<Feature | null>(featurePath, null, {
           maxBackups: DEFAULT_BACKUP_COUNT,
@@ -300,14 +320,21 @@ export class FeatureStateManager {
         if (!feature) continue;
 
         let needsUpdate = false;
+        const originalStatus = feature.status;
 
-        // Reset in_progress features back to ready/backlog
+        // Reset features in active execution states back to a resting state
-        if (feature.status === 'in_progress') {
+        // After a server restart, no processes are actually running
+        const isActiveState =
+          originalStatus === 'in_progress' ||
+          originalStatus === 'interrupted' ||
+          (originalStatus != null && originalStatus.startsWith('pipeline_'));
+
+        if (isActiveState) {
           const hasApprovedPlan = feature.planSpec?.status === 'approved';
           feature.status = hasApprovedPlan ? 'ready' : 'backlog';
           needsUpdate = true;
           logger.info(
-            `[resetStuckFeatures] Reset feature ${feature.id} from in_progress to ${feature.status}`
+            `[${callerLabel}] Reset feature ${feature.id} from ${originalStatus} to ${feature.status}`
           );
         }
 
@@ -316,7 +343,7 @@ export class FeatureStateManager {
           feature.planSpec.status = 'pending';
           needsUpdate = true;
           logger.info(
-            `[resetStuckFeatures] Reset feature ${feature.id} planSpec status from generating to pending`
+            `[${callerLabel}] Reset feature ${feature.id} planSpec status from generating to pending`
           );
         }
 
@@ -327,13 +354,13 @@ export class FeatureStateManager {
               task.status = 'pending';
               needsUpdate = true;
               logger.info(
-                `[resetStuckFeatures] Reset task ${task.id} for feature ${feature.id} from in_progress to pending`
+                `[${callerLabel}] Reset task ${task.id} for feature ${feature.id} from in_progress to pending`
               );
               // Clear currentTaskId if it points to this reverted task
               if (feature.planSpec?.currentTaskId === task.id) {
                 feature.planSpec.currentTaskId = undefined;
                 logger.info(
-                  `[resetStuckFeatures] Cleared planSpec.currentTaskId for feature ${feature.id} (was pointing to reverted task ${task.id})`
+                  `[${callerLabel}] Cleared planSpec.currentTaskId for feature ${feature.id} (was pointing to reverted task ${task.id})`
                 );
               }
             }
@@ -343,19 +370,94 @@ export class FeatureStateManager {
         if (needsUpdate) {
           feature.updatedAt = new Date().toISOString();
           await atomicWriteJson(featurePath, feature, { backupCount: DEFAULT_BACKUP_COUNT });
-          featuresReset++;
+          reconciledCount++;
+          reconciledFeatureIds.push(feature.id);
+          reconciledFeatures.push({
+            id: feature.id,
+            previousStatus: originalStatus,
+            newStatus: feature.status,
+          });
         }
       }
-
-      logger.info(
-        `[resetStuckFeatures] Scanned ${featuresScanned} features, reset ${featuresReset} features for ${projectPath}`
-      );
     } catch (error) {
       // If features directory doesn't exist, that's fine
       if ((error as NodeJS.ErrnoException).code !== 'ENOENT') {
-        logger.error(`[resetStuckFeatures] Error resetting features for ${projectPath}:`, error);
+        logger.error(`[${callerLabel}] Error resetting features for ${projectPath}:`, error);
       }
     }
+
+    return { reconciledFeatures, reconciledFeatureIds, reconciledCount, scanned };
+  }
+
+  /**
+   * Reset features that were stuck in transient states due to server crash.
+   * Called when auto mode is enabled to clean up from previous session.
+   *
+   * Resets:
+   * - in_progress features back to ready (if has plan) or backlog (if no plan)
+   * - interrupted features back to ready (if has plan) or backlog (if no plan)
+   * - pipeline_* features back to ready (if has plan) or backlog (if no plan)
+   * - generating planSpec status back to pending
+   * - in_progress tasks back to pending
+   *
+   * @param projectPath - The project path to reset features for
+   */
+  async resetStuckFeatures(projectPath: string): Promise<void> {
+    const { reconciledCount, scanned } = await this.scanAndResetFeatures(
+      projectPath,
+      'resetStuckFeatures'
+    );
+
+    logger.info(
+      `[resetStuckFeatures] Scanned ${scanned} features, reset ${reconciledCount} features for ${projectPath}`
+    );
+  }
+
+  /**
+   * Reconcile all feature states on server startup.
+   *
+   * This method resets all features stuck in transient states (in_progress,
+   * interrupted, pipeline_*) and emits events so connected UI clients
+   * immediately reflect the corrected states.
+   *
+   * Should be called once during server initialization, before the UI is served,
+   * to ensure feature state consistency after any type of restart (clean, forced, crash).
+   *
+   * @param projectPath - The project path to reconcile features for
+   * @returns The number of features that were reconciled
+   */
+  async reconcileAllFeatureStates(projectPath: string): Promise<number> {
+    logger.info(`[reconcileAllFeatureStates] Starting reconciliation for ${projectPath}`);
+
+    const { reconciledFeatures, reconciledFeatureIds, reconciledCount, scanned } =
+      await this.scanAndResetFeatures(projectPath, 'reconcileAllFeatureStates');
+
+    // Emit per-feature status change events so UI invalidates its cache
+    for (const { id, previousStatus, newStatus } of reconciledFeatures) {
+      this.emitAutoModeEvent('feature_status_changed', {
+        featureId: id,
+        projectPath,
+        status: newStatus,
+        previousStatus,
+        reason: 'server_restart_reconciliation',
+      });
+    }
+
+    // Emit a bulk reconciliation event for the UI
+    if (reconciledCount > 0) {
+      this.emitAutoModeEvent('features_reconciled', {
+        projectPath,
+        reconciledCount,
+        reconciledFeatureIds,
+        message: `Reconciled ${reconciledCount} feature(s) after server restart`,
+      });
+    }
+
+    logger.info(
+      `[reconcileAllFeatureStates] Scanned ${scanned} features, reconciled ${reconciledCount} for ${projectPath}`
+    );
+
+    return reconciledCount;
   }
 
   /**
@@ -532,7 +634,7 @@ export class FeatureStateManager {
    * @param eventType - The event type (e.g., 'auto_mode_summary')
    * @param data - The event payload
    */
-  private emitAutoModeEvent(eventType: string, data: Record<string, unknown>): void {
+  private emitAutoModeEvent(eventType: AutoModeEventType, data: Record<string, unknown>): void {
     // Wrap the event in auto-mode:event format expected by the client
     this.events.emit('auto-mode:event', {
       type: eventType,

apps/server/src/services/typed-event-bus.ts

@@ -40,9 +40,13 @@ export type AutoModeEventType =
   | 'plan_rejected'
   | 'plan_revision_requested'
   | 'plan_revision_warning'
+  | 'plan_spec_updated'
   | 'pipeline_step_started'
   | 'pipeline_step_complete'
-  | string; // Allow other strings for extensibility
+  | 'pipeline_test_failed'
+  | 'pipeline_merge_conflict'
+  | 'feature_status_changed'
+  | 'features_reconciled';
 
 /**
  * TypedEventBus wraps an EventEmitter to provide type-safe event emission

apps/ui/src/hooks/use-query-invalidation.ts

@@ -38,6 +38,8 @@ const FEATURE_LIST_INVALIDATION_EVENTS: AutoModeEvent['type'][] = [
   'plan_rejected',
   'pipeline_step_started',
   'pipeline_step_complete',
+  'feature_status_changed',
+  'features_reconciled',
 ];
 
 /**

apps/ui/src/types/electron.d.ts

@@ -3,7 +3,7 @@
  */
 
 import type { ClaudeUsageResponse, CodexUsageResponse } from '@/store/app-store';
-import type { ParsedTask } from '@automaker/types';
+import type { ParsedTask, FeatureStatusWithPipeline } from '@automaker/types';
 
 export interface ImageAttachment {
   id?: string; // Optional - may not be present in messages loaded from server
@@ -359,6 +359,21 @@ export type AutoModeEvent =
         title?: string;
         status?: string;
       }>;
+    }
+  | {
+      type: 'feature_status_changed';
+      featureId: string;
+      projectPath?: string;
+      status: FeatureStatusWithPipeline;
+      previousStatus: FeatureStatusWithPipeline;
+      reason?: string;
+    }
+  | {
+      type: 'features_reconciled';
+      projectPath?: string;
+      reconciledCount: number;
+      reconciledFeatureIds: string[];
+      message: string;
     };
 
 export type SpecRegenerationEvent =

start-automaker.sh

@@ -36,8 +36,24 @@ elif [[ "$OSTYPE" == "darwin"* ]]; then
 fi
 
 # Port configuration
-DEFAULT_WEB_PORT=3007
+# Defaults can be overridden via AUTOMAKER_WEB_PORT and AUTOMAKER_SERVER_PORT env vars
-DEFAULT_SERVER_PORT=3008
+
+# Validate env-provided ports early (before colors are available)
+if [ -n "$AUTOMAKER_WEB_PORT" ]; then
+    if ! [[ "$AUTOMAKER_WEB_PORT" =~ ^[0-9]+$ ]] || [ "$AUTOMAKER_WEB_PORT" -lt 1 ] || [ "$AUTOMAKER_WEB_PORT" -gt 65535 ]; then
+        echo "Error: AUTOMAKER_WEB_PORT must be a number between 1-65535, got '$AUTOMAKER_WEB_PORT'"
+        exit 1
+    fi
+fi
+if [ -n "$AUTOMAKER_SERVER_PORT" ]; then
+    if ! [[ "$AUTOMAKER_SERVER_PORT" =~ ^[0-9]+$ ]] || [ "$AUTOMAKER_SERVER_PORT" -lt 1 ] || [ "$AUTOMAKER_SERVER_PORT" -gt 65535 ]; then
+        echo "Error: AUTOMAKER_SERVER_PORT must be a number between 1-65535, got '$AUTOMAKER_SERVER_PORT'"
+        exit 1
+    fi
+fi
+
+DEFAULT_WEB_PORT=${AUTOMAKER_WEB_PORT:-3007}
+DEFAULT_SERVER_PORT=${AUTOMAKER_SERVER_PORT:-3008}
 PORT_SEARCH_MAX_ATTEMPTS=100
 WEB_PORT=$DEFAULT_WEB_PORT
 SERVER_PORT=$DEFAULT_SERVER_PORT
@@ -136,6 +152,9 @@ EXAMPLES:
   start-automaker.sh docker       # Launch Docker dev container
   start-automaker.sh --version    # Show version
 
+  AUTOMAKER_WEB_PORT=4000 AUTOMAKER_SERVER_PORT=4001 start-automaker.sh web
+                                  # Launch web mode on custom ports
+
 KEYBOARD SHORTCUTS (in menu):
   Up/Down arrows   Navigate between options
   Enter            Select highlighted option
@@ -146,6 +165,10 @@ HISTORY:
   Your last selected mode is remembered in: ~/.automaker_launcher_history
   Use --no-history to disable this feature
 
+ENVIRONMENT VARIABLES:
+  AUTOMAKER_WEB_PORT     Override default web/UI port (default: 3007)
+  AUTOMAKER_SERVER_PORT  Override default API server port (default: 3008)
+
 PLATFORMS:
   Linux, macOS, Windows (Git Bash, WSL, MSYS2, Cygwin)