deploy-felhom-compose/controller/cmd/controller/main.go

package main

import (
	"context"
	"encoding/base64"
	"encoding/json"
	"flag"
	"fmt"
	"log"
	"net/http"
	"os"
	"os/signal"
	"path/filepath"
	"syscall"
	"time"

	"strings"

	"gitea.dooplex.hu/admin/felhom-controller/internal/api"
	"gitea.dooplex.hu/admin/felhom-controller/internal/backup"
	"gitea.dooplex.hu/admin/felhom-controller/internal/config"
	"gitea.dooplex.hu/admin/felhom-controller/internal/metrics"
	"gitea.dooplex.hu/admin/felhom-controller/internal/monitor"
	"gitea.dooplex.hu/admin/felhom-controller/internal/notify"
	"gitea.dooplex.hu/admin/felhom-controller/internal/report"
	"gitea.dooplex.hu/admin/felhom-controller/internal/scheduler"
	"gitea.dooplex.hu/admin/felhom-controller/internal/selfupdate"
	"gitea.dooplex.hu/admin/felhom-controller/internal/settings"
	"gitea.dooplex.hu/admin/felhom-controller/internal/stacks"
	catalogsync "gitea.dooplex.hu/admin/felhom-controller/internal/sync"
	"gitea.dooplex.hu/admin/felhom-controller/internal/system"
	"gitea.dooplex.hu/admin/felhom-controller/internal/storage"
	"gitea.dooplex.hu/admin/felhom-controller/internal/web"
)

var (
	// Set at build time via ldflags
	Version   = "dev"
	BuildTime = "unknown"
	GitCommit = "unknown"
)

func main() {
	configPath := flag.String("config", "/opt/docker/felhom-controller/controller.yaml", "Path to configuration file")
	showVersion := flag.Bool("version", false, "Show version and exit")
	flag.Parse()

	if *showVersion {
		fmt.Printf("felhom-controller %s (built %s, commit %s)\n", Version, BuildTime, GitCommit)
		os.Exit(0)
	}

	// --- Load configuration ---
	cfg, err := config.Load(*configPath)
	if err != nil {
		log.Fatalf("[FATAL] Failed to load config from %s: %v", *configPath, err)
	}

	logger := setupLogger(cfg)
	logger.Printf("[INFO] felhom-controller %s starting (customer: %s, domain: %s)",
		Version, cfg.Customer.ID, cfg.Customer.Domain)

	// --- Load settings ---
	settingsPath := cfg.Paths.DataDir + "/settings.json"
	sett, err := settings.Load(settingsPath, logger)
	if err != nil {
		logger.Fatalf("[FATAL] Failed to load settings from %s: %v", settingsPath, err)
	}

	// --- Detect fresh deployment (Phase 2+3: DR restore from Hub) ---
	var restorePlan *backup.RestorePlan
	isFreshDeployment := !fileExists(settingsPath)
	if isFreshDeployment && cfg.Hub.Enabled && cfg.Hub.URL != "" {
		logger.Println("[INFO] Fresh deployment detected — checking Hub for infra backup")

		ib, pullErr := report.PullInfraBackup(cfg.Hub.URL, cfg.Hub.APIKey, cfg.Customer.ID)
		if pullErr != nil {
			logger.Printf("[WARN] Could not reach Hub for infra backup: %v", pullErr)
		} else if ib != nil {
			logger.Printf("[INFO] Found infra backup on Hub: %s (%s), %d stacks, synced %s",
				ib.Domain, ib.CustomerID, len(ib.DeployedStacks), ib.Timestamp)

			// Restore settings.json from Hub backup first
			restoreSettingsFromHub(ib, cfg, logger)

			// Re-load settings (now from restored file)
			if restoredSett, loadErr := settings.Load(settingsPath, logger); loadErr == nil {
				sett = restoredSett
				logger.Println("[INFO] Settings reloaded after Hub restore")
			}

			// Restore restic passwords AFTER settings reload so cross-drive password persists
			restorePasswordsFromHub(ib, cfg, sett, logger)

			// Mount drives using stored disk layout
			mountCtx, mountCancel := context.WithTimeout(context.Background(), 2*time.Minute)
			mountedPaths, mountErr := backup.MountDrivesFromLayout(mountCtx, ib.DiskLayout, logger)
			mountCancel()
			if mountErr != nil {
				logger.Printf("[WARN] Drive mounting error: %v", mountErr)
			} else if len(mountedPaths) > 0 {
				logger.Printf("[INFO] Mounted %d drives from Hub disk layout: %v", len(mountedPaths), mountedPaths)
			} else {
				logger.Println("[INFO] No matching drives found to mount from Hub disk layout")
			}

			// Phase 3: Scan mounted drives for backup data and build restore plan
			if len(ib.DeployedStacks) > 0 {
				// Collect mount paths from disk layout
				var drivePaths []string
				for _, dm := range ib.DiskLayout.Mounts {
					if dm.MountPoint != "" {
						drivePaths = append(drivePaths, dm.MountPoint)
					}
				}

				// Convert report stacks to backup scan format
				var infraStacks []backup.InfraStackInfo
				for _, s := range ib.DeployedStacks {
					infraStacks = append(infraStacks, backup.InfraStackInfo{
						Name:        s.Name,
						DisplayName: s.DisplayName,
						HDDPath:     s.HDDPath,
						NeedsHDD:    s.NeedsHDD,
					})
				}

				restorePlan = backup.ScanDrivesForBackups(drivePaths, infraStacks, logger)
				if restorePlan != nil {
					restorePlan.CustomerID = ib.CustomerID
					restorePlan.Domain = ib.Domain
					restorePlan.Timestamp = ib.Timestamp
					logger.Printf("[INFO] DR restore plan ready: %d apps to restore", len(restorePlan.Apps))
				} else {
					logger.Println("[WARN] ScanDrivesForBackups returned nil — no restore plan created")
				}
			}
		} else {
			logger.Println("[INFO] No infra backup found on Hub for this customer")
		}
	}

	// --- Auto-discover storage paths from deployed apps ---
	discoveredPaths := discoverHDDPaths(cfg.Paths.StacksDir, logger)
	sett.AutoDiscoverStoragePaths(discoveredPaths, cfg.Paths.HDDPath, logger)

	// --- Initialize stack manager ---
	stackMgr, err := stacks.NewManager(cfg, logger)
	if err != nil {
		logger.Fatalf("[FATAL] Failed to initialize stack manager: %v", err)
	}

	// Initial stack scan
	if err := stackMgr.ScanStacks(); err != nil {
		logger.Printf("[WARN] Initial stack scan failed: %v", err)
	}

	// Inject missing deploy fields for all deployed stacks on startup
	if names := stackMgr.DeployedStackNames(); len(names) > 0 {
		stackMgr.InjectMissingFields(names)
	}

	// --- Initialize catalog syncer ---
	syncer := catalogsync.New(cfg, logger, stackMgr.ScanStacks, func(updated []string) {
		stackMgr.InjectMissingFields(updated)
	})
	syncer.Start()
	defer syncer.Stop()

	// --- Graceful shutdown context ---
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	// --- Start CPU collector ---
	cpuCollector := system.NewCPUCollector(5 * time.Second)
	cpuCollector.Start(ctx)
	defer cpuCollector.Stop()

	// --- Initialize metrics store + collector ---
	metricsDBPath := "/opt/docker/felhom-controller/data/metrics.db"
	metricsStore, err := metrics.NewMetricsStore(metricsDBPath, logger)
	if err != nil {
		logger.Printf("[WARN] Failed to initialize metrics store: %v — monitoring disabled", err)
	} else {
		logger.Printf("[INFO] Metrics store opened at %s", metricsDBPath)
	}

	if metricsStore != nil {
		defer metricsStore.Close()
		metricsHDDPath := cfg.Paths.HDDPath
		if p := sett.GetDefaultStoragePath(); p != "" {
			metricsHDDPath = p
		}
		metricsCollector := metrics.NewMetricsCollector(metricsStore, cpuCollector, metricsHDDPath, logger)
		metricsCollector.Start(ctx)
		defer metricsCollector.Stop()
		logger.Println("[INFO] Metrics collector started (60s interval)")
	}

	// --- Initialize health pinger (legacy, will be removed) ---
	pinger := monitor.NewPinger(&cfg.Monitoring, logger)

	// Deprecation notice for ping UUIDs
	uuids := cfg.Monitoring.PingUUIDs
	if uuids.Heartbeat != "" || uuids.SystemHealth != "" || uuids.DBDump != "" || uuids.Backup != "" || uuids.BackupIntegrity != "" {
		logger.Println("[INFO] Healthchecks ping UUIDs configured but no longer used — monitoring is now handled by the Hub")
	}

	// --- Initialize backup manager ---
	var backupMgr *backup.Manager
	stackProv := &stackAdapter{
		mgr:             stackMgr,
		getStoragePaths: func() []settings.StoragePath { return sett.GetStoragePaths() },
	}
	if cfg.Backup.Enabled {
		backupMgr = backup.NewManager(cfg, pinger, sett, logger)
		backupMgr.SetStackProvider(stackProv)
		backupMgr.AfterBackup = func() {
			nextDBDump := scheduler.NextDailyRun(cfg.Backup.DBDumpSchedule)
			nextBackup := scheduler.NextDailyRun(cfg.Backup.ResticSchedule)
			backupMgr.RefreshCache(nextDBDump, nextBackup)
		}
		go backupMgr.LoadSnapshotHistory()
	}

	// --- Initialize cross-drive backup runner ---
	crossDriveRunner := backup.NewCrossDriveRunner(sett, stackProv, cfg.Paths.SystemDataPath, cfg.Paths.StacksDir, logger)

	// Wire cross-drive → backup manager for pre-backup DB dumps
	if backupMgr != nil {
		crossDriveRunner.SetDBDumper(backupMgr)
	}

	// --- Initialize alert manager ---
	alertMgr := web.NewAlertManager(logger)

	// --- Initialize notifier ---
	notifier := notify.New(cfg.Hub.URL, cfg.Hub.APIKey, cfg.Customer.ID, sett, logger)

	// --- Initialize self-updater ---
	var updater *selfupdate.Updater
	if cfg.SelfUpdate.Enabled {
		composePath := filepath.Join(filepath.Dir(cfg.Paths.DataDir), "docker-compose.yml")
		updater = selfupdate.NewUpdater(&cfg.SelfUpdate, &cfg.Git, Version, cfg.Paths.DataDir, composePath, logger)
		updater.SetBackupRunningCheck(func() bool {
			return backupMgr != nil && backupMgr.IsRunning()
		})
		// Check for post-update state (did a previous update succeed or fail?)
		if state := updater.VerifyStartup(); state != nil {
			notifier.NotifyControllerUpdated(state.PreviousVersion, state.TargetVersion, state.Status == "success")
		}
		logger.Printf("[INFO] Self-update enabled (check every %s, auto-update: %v, auto-update time: %s)",
			cfg.SelfUpdate.CheckInterval, cfg.SelfUpdate.AutoUpdate, cfg.SelfUpdate.AutoUpdateTime)
	}

	// --- Initialize scheduler ---
	sched := scheduler.New(logger)

	// Existing periodic tasks (migrated from ad-hoc goroutines)
	sched.Every("status-refresh", 30*time.Second, func(ctx context.Context) error {
		return stackMgr.RefreshStatus()
	})
	sched.Every("stack-scan", 2*time.Minute, func(ctx context.Context) error {
		return stackMgr.ScanStacks()
	})

	// Heartbeat — lightweight "I'm alive" signal
	sched.Every("heartbeat", 5*time.Minute, func(ctx context.Context) error {
		pinger.Ping(cfg.Monitoring.PingUUIDs.Heartbeat, "")
		return nil
	})

	// System health ping
	healthInterval, err := time.ParseDuration(cfg.Monitoring.SystemHealthInterval)
	if err != nil {
		healthInterval = 5 * time.Minute
	}
	sched.Every("system-health", healthInterval, func(ctx context.Context) error {
		healthReport := monitor.RunHealthCheck(cfg, cpuCollector, sett.GetStoragePaths())
		body := healthReport.FormatMessage()
		healthUUID := cfg.Monitoring.PingUUIDs.SystemHealth
		if healthReport.Status == "fail" {
			pinger.Fail(healthUUID, body)
		} else {
			pinger.Ping(healthUUID, body)
		}
		// Refresh dashboard alerts from health report
		updateAvailable := false
		latestVersion := ""
		if updater != nil {
			status := updater.GetStatus()
			if status.LastCheck != nil {
				updateAvailable = status.LastCheck.UpdateAvailable
				latestVersion = status.LastCheck.LatestVersion
			}
		}
		alertMgr.Refresh(healthReport, cfg, backupMgr, updateAvailable, latestVersion, sett.GetStoragePaths())
		// Notify on health status changes
		notifier.NotifyHealthChange(healthReport.Status, healthReport.Issues, healthReport.Warnings)
		return nil
	})

	// --- Central hub pusher (declared early so backup closure can reference it) ---
	var hubPusher *report.Pusher
	if cfg.Hub.URL != "" && cfg.Hub.APIKey != "" {
		hubPusher = report.NewPusher(&cfg.Hub, logger)
		// Wire hub push status into alert manager for dashboard alerts
		alertMgr.SetHubPushStatus(func() web.HubPushStatusData {
			s := hubPusher.GetStatus()
			return web.HubPushStatusData{
				LastAttempt: s.LastAttempt,
				LastSuccess: s.LastSuccess,
				LastError:   s.LastError,
				Consecutive: s.Consecutive,
			}
		})
	}

	// Backup daily jobs
	if cfg.Backup.Enabled && backupMgr != nil {
		sched.Daily("db-dump", cfg.Backup.DBDumpSchedule, func(ctx context.Context) error {
			err := backupMgr.RunDBDumps(ctx)
			if err != nil {
				notifier.NotifyDBDumpFailed("Adatbázis mentés sikertelen", err.Error())
			} else {
				notifier.NotifyDBDumpCompleted(notify.DBDumpDetails{})
			}
			return err
		})
		sched.Daily("backup", cfg.Backup.ResticSchedule, func(ctx context.Context) error {
			err := backupMgr.RunBackup(ctx)
			if err != nil {
				notifier.NotifyBackupFailed("Biztonsági mentés sikertelen", err.Error())
			} else {
				notifier.NotifyBackupCompleted(notify.BackupDetails{})
			}
			// Phase 3: Chain cross-drive backups immediately after restic (regardless of restic success)
			// Daily jobs run every night; weekly jobs only on Sunday
			if crossDriveRunner != nil {
				if cdErr := crossDriveRunner.RunAllScheduled(ctx, "daily"); cdErr != nil {
					logger.Printf("[WARN] Cross-drive daily backup had errors: %v", cdErr)
				}
				if time.Now().Weekday() == time.Sunday {
					if cdErr := crossDriveRunner.RunAllScheduled(ctx, "weekly"); cdErr != nil {
						logger.Printf("[WARN] Cross-drive weekly backup had errors: %v", cdErr)
					}
				}
			}
			// Push infra backup to Hub after all backup tiers complete
			if hubPusher != nil && cfg.Hub.Enabled {
				go pushInfraBackup(cfg, sett, stackProv, hubPusher, logger)
			}
			return err
		})

		// Weekly integrity check — Sunday 04:00
		sched.Daily("backup-integrity", "04:00", func(ctx context.Context) error {
			if time.Now().Weekday() != time.Sunday {
				return nil
			}
			err := backupMgr.RunIntegrityCheck(ctx)
			if err != nil {
				notifier.NotifyIntegrityFailed("Mentés integritás ellenőrzés sikertelen", err.Error())
			} else {
				notifier.NotifyIntegrityOK("Mentés integritás ellenőrzés sikeres")
			}
			return err
		})

		// Cache refresh: every 5 minutes
		sched.Every("backup-cache", 5*time.Minute, func(ctx context.Context) error {
			nextDBDump := scheduler.NextDailyRun(cfg.Backup.DBDumpSchedule)
			nextBackup := scheduler.NextDailyRun(cfg.Backup.ResticSchedule)
			backupMgr.RefreshCache(nextDBDump, nextBackup)
			return nil
		})
	}

	// Metrics prune — daily at 04:00
	if metricsStore != nil {
		sched.Daily("metrics-prune", "04:00", func(ctx context.Context) error {
			deleted, err := metricsStore.Prune(30 * 24 * time.Hour)
			if err != nil {
				return err
			}
			logger.Printf("[INFO] Pruned %d old metric rows", deleted)
			return nil
		})
	}

	// --- Central hub reporting schedule ---
	if hubPusher != nil {
		if cfg.Hub.Enabled {
			pushInterval, err := time.ParseDuration(cfg.Hub.PushInterval)
			if err != nil {
				pushInterval = 15 * time.Minute
			}
			sched.Every("hub-report", pushInterval, func(ctx context.Context) error {
				r := report.BuildReport(cfg, *configPath, stackMgr, backupMgr, cpuCollector, metricsStore, Version, sett.GetStoragePaths())
				return hubPusher.Push(r)
			})
			logger.Printf("[INFO] Hub reporting enabled (every %s to %s)", pushInterval, cfg.Hub.URL)
		} else {
			logger.Printf("[INFO] Hub reporting disabled — will send disabled notification to %s", cfg.Hub.URL)
		}
	}

	// Self-update scheduler jobs
	if cfg.SelfUpdate.Enabled && updater != nil {
		// Periodic version check (populates UI, never triggers update)
		checkInterval, ciErr := time.ParseDuration(cfg.SelfUpdate.CheckInterval)
		if ciErr != nil {
			checkInterval = 6 * time.Hour
		}
		sched.Every("selfupdate-check", checkInterval, func(ctx context.Context) error {
			result := updater.CheckForUpdate()
			if result.UpdateAvailable {
				logger.Printf("[INFO] Update available: %s -> %s", result.CurrentVersion, result.LatestVersion)
			}
			return nil
		})

		// Auto-update (daily, fires after typical backup completion)
		if cfg.SelfUpdate.AutoUpdate {
			sched.Daily("selfupdate-auto", cfg.SelfUpdate.AutoUpdateTime, func(ctx context.Context) error {
				result := updater.CheckForUpdate()
				if !result.UpdateAvailable {
					return nil
				}
				if err := updater.TriggerUpdate("auto"); err != nil {
					logger.Printf("[WARN] Auto-update skipped: %v", err)
				}
				return nil
			})
		}
	}

	// --- Storage watchdog ---
	storageWatchdog := monitor.NewStorageWatchdog(sett, &watchdogStackAdapter{mgr: stackMgr}, notifier, cfg, logger)
	storageWatchdog.SetAlertRefresh(func() {
		healthReport := monitor.RunHealthCheck(cfg, cpuCollector, sett.GetStoragePaths())
		updateAvailable := false
		latestVersion := ""
		if updater != nil {
			status := updater.GetStatus()
			if status.LastCheck != nil {
				updateAvailable = status.LastCheck.UpdateAvailable
				latestVersion = status.LastCheck.LatestVersion
			}
		}
		alertMgr.Refresh(healthReport, cfg, backupMgr, updateAvailable, latestVersion, sett.GetStoragePaths())
	})
	if hubPusher != nil {
		storageWatchdog.SetHubReportPusher(func() {
			r := report.BuildReport(cfg, *configPath, stackMgr, backupMgr, cpuCollector, metricsStore, Version, sett.GetStoragePaths())
			hubPusher.Push(r)
		})
	}
	if backupMgr != nil {
		storageWatchdog.SetRepoUnlocker(func(ctx context.Context, repoPath string) error {
			return backupMgr.UnlockRepo(ctx, repoPath)
		})
	}
	sched.Every("storage-watchdog", 5*time.Second, func(ctx context.Context) error {
		return storageWatchdog.Check(ctx)
	})

	sched.Start(ctx)
	defer sched.Stop()

	// Fire startup pings + hub report immediately (don't wait for first scheduler tick)
	go func() {
		time.Sleep(5 * time.Second) // Let all subsystems fully initialize

		// Push controller startup event to Hub
		notifier.NotifyControllerStarted(Version)

		// Heartbeat ping
		pinger.Ping(cfg.Monitoring.PingUUIDs.Heartbeat, "startup")
		logger.Println("[INFO] Startup heartbeat ping sent")

		// System health ping
		healthReport := monitor.RunHealthCheck(cfg, cpuCollector, sett.GetStoragePaths())
		body := healthReport.FormatMessage()
		healthUUID := cfg.Monitoring.PingUUIDs.SystemHealth
		if healthReport.Status == "fail" {
			pinger.Fail(healthUUID, body)
		} else {
			pinger.Ping(healthUUID, body)
		}
		logger.Printf("[INFO] Startup health ping sent (status: %s)", healthReport.Status)

		// Hub report
		if hubPusher != nil {
			if cfg.Hub.Enabled {
				r := report.BuildReport(cfg, *configPath, stackMgr, backupMgr, cpuCollector, metricsStore, Version, sett.GetStoragePaths())
				var pushErr error
				for attempt := 1; attempt <= 3; attempt++ {
					pushErr = hubPusher.Push(r)
					if pushErr == nil {
						logger.Println("[INFO] Startup hub report sent")
						break
					}
					logger.Printf("[WARN] Startup hub report attempt %d/3 failed: %v", attempt, pushErr)
					if attempt < 3 {
						time.Sleep(15 * time.Second)
					}
				}
				if pushErr != nil {
					logger.Printf("[WARN] Startup hub report failed after 3 attempts — next scheduled push in %s", cfg.Hub.PushInterval)
				}
				// Also push infra backup on startup
				go pushInfraBackup(cfg, sett, stackProv, hubPusher, logger)
			} else {
				// Send a minimal "disabled" notification so hub knows reporting is intentionally off
				r := &report.Report{
					Version:           1,
					CustomerID:        cfg.Customer.ID,
					CustomerName:      cfg.Customer.Name,
					ControllerVersion: Version,
					Timestamp:         time.Now().UTC(),
					ReportingDisabled: true,
					Health:            report.HealthReport{Status: "disabled", Issues: []string{}, Warnings: []string{}},
					Stacks:            report.StacksReport{Deployed: []string{}, Available: []string{}},
					Containers:        report.ContainerReport{List: []report.ContainerDetailReport{}},
				}
				hubPusher.PushOnce(r)
			}
		}

		// Initial self-update check (so settings page shows version info quickly)
		if updater != nil {
			time.Sleep(25 * time.Second) // Additional delay after hub report
			result := updater.CheckForUpdate()
			if result.UpdateAvailable {
				logger.Printf("[INFO] Startup: update available %s -> %s", result.CurrentVersion, result.LatestVersion)
			} else if result.Error != "" {
				logger.Printf("[DEBUG] Startup version check: %s", result.Error)
			}
		}
	}()

	// Initial backup cache population (don't block startup)
	if cfg.Backup.Enabled && backupMgr != nil {
		go func() {
			nextDBDump := scheduler.NextDailyRun(cfg.Backup.DBDumpSchedule)
			nextBackup := scheduler.NextDailyRun(cfg.Backup.ResticSchedule)
			backupMgr.RefreshCache(nextDBDump, nextBackup)
		}()
	}

	// Sync notification preferences to hub on startup (handles hub DB rebuild recovery)
	if notifier.IsEnabled() {
		go func() {
			prefs := sett.GetNotificationPrefs()
			if prefs.Email != "" {
				if err := notifier.SyncPreferences(prefs.Email, prefs.EnabledEvents, prefs.CooldownHours); err != nil {
					logger.Printf("[WARN] Failed to sync notification preferences on startup: %v", err)
				}
			}
		}()
	}

	// Initial alert refresh (so alerts appear immediately, not after first 5min health check)
	go func() {
		report := monitor.RunHealthCheck(cfg, cpuCollector, sett.GetStoragePaths())
		alertMgr.Refresh(report, cfg, backupMgr, false, "")
	}()

	// --- Initialize API router ---
	apiRouter := api.NewRouter(cfg, *configPath, sett, stackMgr, syncer, cpuCollector, backupMgr, crossDriveRunner, metricsStore, updater, notifier, logger)
	if hubPusher != nil {
		apiRouter.OnConfigApplied = func() {
			pushInfraBackup(cfg, sett, stackProv, hubPusher, logger)
		}
	}

	// --- Initialize web server ---
	webServer := web.NewServer(cfg, stackMgr, cpuCollector, backupMgr, crossDriveRunner, sched, sett, alertMgr, notifier, updater, logger, Version)
	webServer.SetStorageWatchdog(storageWatchdog)
	if hubPusher != nil {
		webServer.SetHubPushStatus(func() web.HubPushStatusData {
			s := hubPusher.GetStatus()
			return web.HubPushStatusData{
				LastAttempt: s.LastAttempt,
				LastSuccess: s.LastSuccess,
				LastError:   s.LastError,
				Consecutive: s.Consecutive,
			}
		})
	}

	// --- Initialize drive migrator ---
	driveMigrator := &storage.DriveMigrator{
		Sett:          sett,
		StackProvider: &driveMigrateStackAdapter{mgr: stackMgr},
		Logger:        logger,
	}
	// Only set BackupTrigger if backup is enabled (avoid non-nil interface with nil concrete value)
	if backupMgr != nil {
		driveMigrator.BackupTrigger = backupMgr
	}
	driveMigrator.AlertRefresh = func() {
		healthReport := monitor.RunHealthCheck(cfg, cpuCollector, sett.GetStoragePaths())
		updateAvailable := false
		latestVersion := ""
		if updater != nil {
			status := updater.GetStatus()
			if status.LastCheck != nil {
				updateAvailable = status.LastCheck.UpdateAvailable
				latestVersion = status.LastCheck.LatestVersion
			}
		}
		alertMgr.Refresh(healthReport, cfg, backupMgr, updateAvailable, latestVersion, sett.GetStoragePaths())
	}
	if hubPusher != nil {
		driveMigrator.PushHubReport = func() {
			r := report.BuildReport(cfg, *configPath, stackMgr, backupMgr, cpuCollector, metricsStore, Version, sett.GetStoragePaths())
			hubPusher.Push(r)
		}
		driveMigrator.PushInfraBackup = func() {
			pushInfraBackup(cfg, sett, stackProv, hubPusher, logger)
		}
	}
	driveMigrator.SyncFBMounts = func() {
		webServer.SyncFileBrowserMounts()
	}
	webServer.SetDriveMigrator(driveMigrator)

	// Wire migration-active check into backup manager
	if backupMgr != nil {
		backupMgr.MigrationActiveCheck = driveMigrator.IsActive
	}

	// Phase 3: Set DR restore mode if a restore plan was built
	if restorePlan != nil && len(restorePlan.Apps) > 0 {
		webServer.SetRestoreState(restorePlan)
		logger.Println("[INFO] DR restore mode activated — all web routes redirect to /restore")
	}

	// --- Build HTTP mux ---
	mux := http.NewServeMux()

	// API routes (no auth for health endpoint, auth for everything else)
	mux.HandleFunc("/api/health", apiRouter.HealthHandler)
	// Storage API routes handled by web server (longer prefix takes precedence over /api/)
	mux.Handle("/api/storage/", webServer.RequireAuth(http.HandlerFunc(webServer.ServeStorageAPI)))
	// Self-update API — accepts session auth OR hub API key (for external triggering)
	mux.Handle("/api/selfupdate/", selfUpdateAuthMiddleware(cfg, webServer, http.HandlerFunc(apiRouter.ServeHTTP)))
	// Config API — accepts session auth OR hub API key (for Hub config push)
	mux.Handle("/api/config/", selfUpdateAuthMiddleware(cfg, webServer, http.HandlerFunc(apiRouter.ServeHTTP)))
	mux.Handle("/api/", webServer.RequireAuth(http.HandlerFunc(apiRouter.ServeHTTP)))

	// Web UI routes (auth required)
	mux.Handle("/", webServer.RequireAuth(http.HandlerFunc(webServer.ServeHTTP)))

	// --- Start HTTP server ---
	server := &http.Server{
		Addr:         cfg.Web.Listen,
		Handler:      mux,
		ReadTimeout:  30 * time.Second,
		WriteTimeout: 60 * time.Second,
		IdleTimeout:  120 * time.Second,
	}

	sigCh := make(chan os.Signal, 1)
	signal.Notify(sigCh, syscall.SIGINT, syscall.SIGTERM)

	go func() {
		sig := <-sigCh
		logger.Printf("[INFO] Received signal %v, shutting down...", sig)
		cancel()

		shutdownCtx, shutdownCancel := context.WithTimeout(context.Background(), 15*time.Second)
		defer shutdownCancel()

		if err := server.Shutdown(shutdownCtx); err != nil {
			logger.Printf("[ERROR] HTTP server shutdown error: %v", err)
		}
	}()

	logger.Printf("[INFO] Web UI listening on %s", cfg.Web.Listen)
	if err := server.ListenAndServe(); err != http.ErrServerClosed {
		logger.Fatalf("[FATAL] HTTP server error: %v", err)
	}

	logger.Println("[INFO] felhom-controller stopped")
}

// selfUpdateAuthMiddleware allows access via session auth (normal UI) OR hub API key bearer token (external).
func selfUpdateAuthMiddleware(cfg *config.Config, webServer *web.Server, next http.Handler) http.Handler {
	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
		// Check bearer token first (for external API calls: hub, build scripts)
		if auth := r.Header.Get("Authorization"); strings.HasPrefix(auth, "Bearer ") {
			token := strings.TrimPrefix(auth, "Bearer ")
			if token != "" && cfg.Hub.APIKey != "" && token == cfg.Hub.APIKey {
				next.ServeHTTP(w, r)
				return
			}
		}
		// Fall back to session auth
		webServer.RequireAuth(next).ServeHTTP(w, r)
	})
}

func setupLogger(cfg *config.Config) *log.Logger {
	// For now, log to stdout. File logging will be added later.
	logger := log.New(os.Stdout, "", log.LstdFlags)

	if cfg.Logging.Level == "debug" {
		logger.SetFlags(log.LstdFlags | log.Lshortfile)
	}

	return logger
}

// stackAdapter implements backup.StackDataProvider using stacks.Manager.
type stackAdapter struct {
	mgr             *stacks.Manager
	getStoragePaths func() []settings.StoragePath
}

func (a *stackAdapter) GetStackComposePath(name string) (string, bool) {
	s, ok := a.mgr.GetStack(name)
	if !ok {
		return "", false
	}
	return s.ComposePath, true
}

func (a *stackAdapter) ListDeployedStacks() []backup.StackSummary {
	var result []backup.StackSummary
	for _, s := range a.mgr.GetStacks() {
		if !s.Deployed {
			continue
		}
		result = append(result, backup.StackSummary{
			Name:        s.Name,
			DisplayName: s.Meta.DisplayName,
			ComposePath: s.ComposePath,
			NeedsHDD:    s.Meta.Resources.NeedsHDD,
		})
	}
	return result
}

func (a *stackAdapter) StopStack(name string) error {
	return a.mgr.StopStack(name)
}

func (a *stackAdapter) StartStack(name string) error {
	return a.mgr.StartStack(name)
}

func (a *stackAdapter) GetStackHDDMounts(name string) []string {
	s, ok := a.mgr.GetStack(name)
	if !ok {
		return nil
	}

	// Priority 1: Read the app's own HDD_PATH from its app.yaml
	stackDir := filepath.Dir(s.ComposePath)
	appCfg := stacks.LoadAppConfig(stackDir)
	if appCfg != nil && appCfg.Env["HDD_PATH"] != "" {
		return stacks.ParseComposeHDDMounts(s.ComposePath, appCfg.Env["HDD_PATH"])
	}

	// Priority 2: Try all registered storage paths (fallback)
	var allMounts []string
	seen := make(map[string]bool)
	for _, sp := range a.getStoragePaths() {
		mounts := stacks.ParseComposeHDDMounts(s.ComposePath, sp.Path)
		for _, m := range mounts {
			if !seen[m] {
				seen[m] = true
				allMounts = append(allMounts, m)
			}
		}
	}
	return allMounts
}

func (a *stackAdapter) GetStackHDDPath(name string) string {
	s, ok := a.mgr.GetStack(name)
	if !ok {
		return ""
	}
	stackDir := filepath.Dir(s.ComposePath)
	appCfg := stacks.LoadAppConfig(stackDir)
	if appCfg != nil && appCfg.Env["HDD_PATH"] != "" {
		return filepath.Clean(appCfg.Env["HDD_PATH"])
	}
	return ""
}

// watchdogStackAdapter implements monitor.WatchdogStackProvider using stacks.Manager.
type watchdogStackAdapter struct {
	mgr *stacks.Manager
}

func (a *watchdogStackAdapter) ListDeployedStacks() []monitor.WatchdogStackInfo {
	var result []monitor.WatchdogStackInfo
	for _, s := range a.mgr.GetStacks() {
		if !s.Deployed {
			continue
		}
		result = append(result, monitor.WatchdogStackInfo{Name: s.Name})
	}
	return result
}

func (a *watchdogStackAdapter) GetStackHDDPath(name string) string {
	s, ok := a.mgr.GetStack(name)
	if !ok {
		return ""
	}
	stackDir := filepath.Dir(s.ComposePath)
	appCfg := stacks.LoadAppConfig(stackDir)
	if appCfg != nil && appCfg.Env["HDD_PATH"] != "" {
		return filepath.Clean(appCfg.Env["HDD_PATH"])
	}
	return ""
}

func (a *watchdogStackAdapter) StopStack(name string) error {
	return a.mgr.StopStack(name)
}

func (a *watchdogStackAdapter) StartStack(name string) error {
	return a.mgr.StartStack(name)
}

// driveMigrateStackAdapter implements storage.StackProviderForMigration using stacks.Manager.
type driveMigrateStackAdapter struct {
	mgr *stacks.Manager
}

func (a *driveMigrateStackAdapter) ListDeployedStacks() []storage.StackSummaryForMigration {
	var result []storage.StackSummaryForMigration
	for _, s := range a.mgr.GetStacks() {
		if !s.Deployed {
			continue
		}
		result = append(result, storage.StackSummaryForMigration{
			Name:        s.Name,
			DisplayName: s.Meta.DisplayName,
		})
	}
	return result
}

func (a *driveMigrateStackAdapter) GetStackHDDPath(name string) string {
	s, ok := a.mgr.GetStack(name)
	if !ok {
		return ""
	}
	stackDir := filepath.Dir(s.ComposePath)
	appCfg := stacks.LoadAppConfig(stackDir)
	if appCfg != nil && appCfg.Env["HDD_PATH"] != "" {
		return filepath.Clean(appCfg.Env["HDD_PATH"])
	}
	return ""
}

func (a *driveMigrateStackAdapter) StopStack(name string) error {
	return a.mgr.StopStack(name)
}

func (a *driveMigrateStackAdapter) StartStack(name string) error {
	return a.mgr.StartStack(name)
}

func (a *driveMigrateStackAdapter) UpdateStackHDDPath(name, newPath string) error {
	s, ok := a.mgr.GetStack(name)
	if !ok {
		return fmt.Errorf("stack not found: %s", name)
	}
	stackDir := filepath.Dir(s.ComposePath)
	appCfg := stacks.LoadAppConfig(stackDir)
	if appCfg == nil {
		return fmt.Errorf("app.yaml not found for stack: %s", name)
	}
	appCfg.Env["HDD_PATH"] = newPath
	return stacks.SaveAppConfig(stackDir, appCfg)
}

func (a *driveMigrateStackAdapter) StackExists(name string) bool {
	_, ok := a.mgr.GetStack(name)
	return ok
}

// pushInfraBackup builds and sends the infrastructure snapshot to the Hub.
func pushInfraBackup(cfg *config.Config, sett *settings.Settings,
	stackProv *stackAdapter, pusher *report.Pusher, logger *log.Logger) {

	ib, err := report.BuildInfraBackup(
		cfg.Customer.ID, cfg.Customer.Domain, Version,
		"/opt/docker/felhom-controller/controller.yaml",
		filepath.Join(cfg.Paths.DataDir, "settings.json"),
		cfg.Backup.ResticPasswordFile,
		cfg.Paths.SystemDataPath,
		sett, stackProv, logger,
	)
	if err != nil {
		logger.Printf("[WARN] Failed to build infra backup: %v", err)
		return
	}

	data, err := json.Marshal(ib)
	if err != nil {
		logger.Printf("[WARN] Failed to marshal infra backup: %v", err)
		return
	}

	if err := pusher.PushInfraBackup(data); err != nil {
		logger.Printf("[WARN] Failed to push infra backup to Hub: %v", err)
	}
}

// fileExists returns true if the path exists (file or directory).
func fileExists(path string) bool {
	_, err := os.Stat(path)
	return err == nil
}

// restorePasswordsFromHub restores restic passwords from a Hub infra backup.
func restorePasswordsFromHub(ib *report.InfraBackup, cfg *config.Config,
	sett *settings.Settings, logger *log.Logger) {

	if ib.ResticPassword != "" {
		decoded, err := base64.StdEncoding.DecodeString(ib.ResticPassword)
		if err == nil && len(decoded) > 0 {
			dir := filepath.Dir(cfg.Backup.ResticPasswordFile)
			if err := os.MkdirAll(dir, 0700); err != nil {
				logger.Printf("[WARN] Failed to create restic password directory %s: %v", dir, err)
			} else if err := os.WriteFile(cfg.Backup.ResticPasswordFile, decoded, 0600); err == nil {
				logger.Println("[INFO] Primary restic password restored from Hub")
			} else {
				logger.Printf("[WARN] Failed to write restic password file: %v", err)
			}
		}
	}

}

// restoreSettingsFromHub restores settings.json from a Hub infra backup.
func restoreSettingsFromHub(ib *report.InfraBackup, cfg *config.Config, logger *log.Logger) {
	if ib.SettingsJSONB64 == "" {
		return
	}
	decoded, err := base64.StdEncoding.DecodeString(ib.SettingsJSONB64)
	if err != nil {
		logger.Printf("[WARN] Failed to decode settings from Hub: %v", err)
		return
	}
	if err := os.MkdirAll(cfg.Paths.DataDir, 0755); err != nil {
		logger.Printf("[WARN] Failed to create data directory for settings restore: %v", err)
		return
	}
	settingsPath := filepath.Join(cfg.Paths.DataDir, "settings.json")
	if err := os.WriteFile(settingsPath, decoded, 0600); err != nil {
		logger.Printf("[WARN] Failed to write restored settings.json: %v", err)
	} else {
		logger.Println("[INFO] Settings restored from Hub backup")
	}
}

// discoverHDDPaths scans deployed apps' app.yaml for HDD_PATH env values.
func discoverHDDPaths(stacksDir string, logger *log.Logger) []string {
	entries, err := os.ReadDir(stacksDir)
	if err != nil {
		logger.Printf("[WARN] Cannot read stacks dir for HDD path discovery: %v", err)
		return nil
	}
	seen := make(map[string]bool)
	var paths []string
	for _, e := range entries {
		if !e.IsDir() {
			continue
		}
		appCfg := stacks.LoadAppConfig(filepath.Join(stacksDir, e.Name()))
		if appCfg == nil || !appCfg.Deployed {
			continue
		}
		if hddPath, ok := appCfg.Env["HDD_PATH"]; ok && hddPath != "" {
			cleaned := filepath.Clean(hddPath)
			if !seen[cleaned] {
				seen[cleaned] = true
				paths = append(paths, cleaned)
			}
		}
	}
	return paths
}