Fix autoupdate, add healtcheck status to the dashboard

internal/worker/handlers.go

@@ -751,6 +751,110 @@ func (s *Service) handleUpdateStatus(w http.ResponseWriter, r *http.Request) {
 	writeJSON(w, status)
 }
 
+// ComponentHealth represents the health status of a single component.
+type ComponentHealth struct {
+	Name    string `json:"name"`
+	Status  string `json:"status"` // "healthy", "degraded", "unhealthy"
+	Message string `json:"message,omitempty"`
+}
+
+// SelfCheckResponse contains the health status of all components.
+type SelfCheckResponse struct {
+	Overall    string            `json:"overall"` // "healthy", "degraded", "unhealthy"
+	Version    string            `json:"version"`
+	Uptime     string            `json:"uptime"`
+	Components []ComponentHealth `json:"components"`
+}
+
+// handleSelfCheck returns the health status of all components.
+func (s *Service) handleSelfCheck(w http.ResponseWriter, r *http.Request) {
+	components := []ComponentHealth{}
+	overall := "healthy"
+
+	// Check Worker Service
+	workerStatus := ComponentHealth{Name: "Worker Service", Status: "healthy"}
+	if !s.ready.Load() {
+		if err := s.GetInitError(); err != nil {
+			workerStatus.Status = "unhealthy"
+			workerStatus.Message = err.Error()
+			overall = "unhealthy"
+		} else {
+			workerStatus.Status = "degraded"
+			workerStatus.Message = "Initializing"
+			if overall == "healthy" {
+				overall = "degraded"
+			}
+		}
+	}
+	components = append(components, workerStatus)
+
+	// Check SQLite Database
+	dbStatus := ComponentHealth{Name: "SQLite Database", Status: "healthy"}
+	if s.store == nil {
+		dbStatus.Status = "unhealthy"
+		dbStatus.Message = "Not initialized"
+		overall = "unhealthy"
+	} else if err := s.store.Ping(); err != nil {
+		dbStatus.Status = "unhealthy"
+		dbStatus.Message = err.Error()
+		overall = "unhealthy"
+	}
+	components = append(components, dbStatus)
+
+	// Check ChromaDB
+	chromaStatus := ComponentHealth{Name: "ChromaDB", Status: "healthy"}
+	if s.chromaClient == nil {
+		chromaStatus.Status = "degraded"
+		chromaStatus.Message = "Not configured"
+		if overall == "healthy" {
+			overall = "degraded"
+		}
+	} else if !s.chromaClient.IsConnected() {
+		chromaStatus.Status = "degraded"
+		chromaStatus.Message = "Not connected"
+		if overall == "healthy" {
+			overall = "degraded"
+		}
+	}
+	components = append(components, chromaStatus)
+
+	// Check SDK Processor
+	sdkStatus := ComponentHealth{Name: "SDK Processor", Status: "healthy"}
+	if s.processor == nil {
+		sdkStatus.Status = "degraded"
+		sdkStatus.Message = "Not initialized"
+		if overall == "healthy" {
+			overall = "degraded"
+		}
+	} else if !s.processor.IsAvailable() {
+		sdkStatus.Status = "degraded"
+		sdkStatus.Message = "Claude CLI not available"
+		if overall == "healthy" {
+			overall = "degraded"
+		}
+	}
+	components = append(components, sdkStatus)
+
+	// Check SSE Broadcaster
+	sseStatus := ComponentHealth{Name: "SSE Broadcaster", Status: "healthy"}
+	if s.sseBroadcaster == nil {
+		sseStatus.Status = "unhealthy"
+		sseStatus.Message = "Not initialized"
+		overall = "unhealthy"
+	}
+	components = append(components, sseStatus)
+
+	// Calculate uptime
+	uptime := time.Since(s.startTime).Round(time.Second).String()
+
+	writeJSON(w, SelfCheckResponse{
+		Overall:    overall,
+		Version:    s.version,
+		Uptime:     uptime,
+		Components: components,
+	})
+}
+
 // handleUpdateRestart restarts the worker with the new binary.
 func (s *Service) handleUpdateRestart(w http.ResponseWriter, r *http.Request) {
 	status := s.updater.GetStatus()

internal/worker/sdk/processor.go

@@ -92,6 +92,12 @@ func NewProcessor(observationStore *sqlite.ObservationStore, summaryStore *sqlit
 	}, nil
 }
 
+// IsAvailable checks if the Claude CLI is available for processing.
+func (p *Processor) IsAvailable() bool {
+	_, err := os.Stat(p.claudePath)
+	return err == nil
+}
+
 // ProcessObservation processes a single tool observation and extracts insights.
 func (p *Processor) ProcessObservation(ctx context.Context, sdkSessionID, project string, toolName string, toolInput, toolResponse interface{}, promptNumber int, cwd string) error {
 	p.mu.Lock()

internal/worker/service.go

@@ -602,6 +602,9 @@ func (s *Service) setupRoutes() {
 	s.router.Get("/api/update/status", s.handleUpdateStatus)
 	s.router.Post("/api/update/restart", s.handleUpdateRestart)
 
+	// Selfcheck endpoint (works before DB is ready - checks all components)
+	s.router.Get("/api/selfcheck", s.handleSelfCheck)
+
 	// SSE endpoint (works before DB is ready)
 	s.router.Get("/api/events", s.sseBroadcaster.HandleSSE)
 
@@ -668,11 +671,34 @@ func (s *Service) Start() error {
 		ReadHeaderTimeout: 10 * time.Second,
 	}
 
+	// Check if we're in restart mode (after update)
+	isRestart := os.Getenv("CLAUDE_MNEMONIC_RESTART") == "1"
+
 	s.wg.Add(1)
 	go func() {
 		defer s.wg.Done()
-		if err := s.server.ListenAndServe(); err != http.ErrServerClosed {
-			log.Error().Err(err).Msg("HTTP server error")
+
+		var lastErr error
+		maxRetries := 1
+		if isRestart {
+			maxRetries = 10 // Retry up to 10 times during restart
+		}
+
+		for i := 0; i < maxRetries; i++ {
+			lastErr = s.server.ListenAndServe()
+			if lastErr == http.ErrServerClosed {
+				return // Normal shutdown
+			}
+
+			if i < maxRetries-1 && isRestart {
+				log.Warn().Err(lastErr).Int("retry", i+1).Msg("Port not ready, retrying...")
+				time.Sleep(500 * time.Millisecond)
+				continue
+			}
+		}
+
+		if lastErr != nil {
+			log.Error().Err(lastErr).Msg("HTTP server error")
 		}
 	}()
 
@@ -681,6 +707,7 @@ func (s *Service) Start() error {
 	log.Info().
 		Int("port", port).
 		Int("pid", getPID()).
+		Bool("restart_mode", isRestart).
 		Msg("Worker HTTP server started (initialization in progress)")
 
 	return nil

internal/worker/static/placeholder.html

@@ -1 +0,0 @@
-<!-- Placeholder for go:embed - replaced by UI build -->