Files
deploy-felhom-compose/controller/internal/backup/restore.go
T
admin be7803c0ac v0.24.0 — Pre-testing observability: debug logging, diagnostic dump, startup self-test
- Add [DEBUG] logging across all modules (backup, storage, sync, selfupdate,
  monitor, notify, report, assets, setup) gated behind logging.level: "debug"
- Add /api/debug/dump endpoint returning full controller state JSON (debug only)
- Add startup self-test validating 9 subsystems (Docker, dirs, storage, hub,
  restic repos, metrics DB) with pass/warn/fail summary
- New packages: internal/selftest, internal/util
- Constructor/signature changes: debug bool params, logger params on
  RunHealthCheck and BuildReport, smart watchdog probe logging

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-21 18:32:26 +01:00

134 lines
4.2 KiB
Go

package backup
import (
"fmt"
"path/filepath"
"regexp"
)
// snapshotIDRe validates restic snapshot IDs: 8-64 lowercase hex characters.
var snapshotIDRe = regexp.MustCompile(`^[0-9a-f]{8,64}$`)
// RestoreApp restores an app from a restic snapshot.
// All apps get config + DB dump restored. Apps with HDD data also get user data restored.
func (m *Manager) RestoreApp(stackName, snapshotID string) error {
if m.stackProvider == nil {
return fmt.Errorf("stack provider not configured")
}
// Validate snapshot ID format
if !snapshotIDRe.MatchString(snapshotID) {
return fmt.Errorf("invalid snapshot ID: must be 8-64 lowercase hex characters")
}
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: stack=%s, snapshotID=%s", stackName, snapshotID)
}
// Prevent concurrent operations
m.mu.Lock()
if m.running {
m.mu.Unlock()
return fmt.Errorf("backup or restore already in progress")
}
m.running = true
m.mu.Unlock()
defer func() {
m.mu.Lock()
m.running = false
m.mu.Unlock()
}()
// Determine what to restore
hddMounts := m.stackProvider.GetStackHDDMounts(stackName)
hasHDD := len(hddMounts) > 0
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: %s has %d HDD mount(s), hasHDD=%v", stackName, len(hddMounts), hasHDD)
}
// Build list of paths to restore from the snapshot
var restorePaths []string
// Always restore the stack's config dir (compose + app.yaml + .felhom.yml)
composePath, ok := m.stackProvider.GetStackComposePath(stackName)
if ok {
stackDir := filepath.Dir(composePath)
restorePaths = append(restorePaths, stackDir)
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: will restore config dir: %s", stackDir)
}
}
// Restore DB dump files for this stack (per-drive path)
drivePath := m.GetAppDrivePath(stackName)
dumpDir := AppDBDumpPath(drivePath, stackName)
restorePaths = append(restorePaths, dumpDir)
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: will restore DB dump dir: %s", dumpDir)
}
// Restore HDD data (always included for apps that have it — backup is mandatory)
if hasHDD {
restorePaths = append(restorePaths, hddMounts...)
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: will restore HDD data: %v", hddMounts)
}
}
if len(restorePaths) == 0 {
return fmt.Errorf("no restorable paths found for %s", stackName)
}
// Use the app's primary restic repo
repoPath := PrimaryResticRepoPath(drivePath)
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: using repo=%s, %d restore path(s)", repoPath, len(restorePaths))
}
m.logger.Printf("[WARN] RESTORE starting: stack=%s, snapshot=%s, repo=%s, paths=%v, hasHDD=%v",
stackName, snapshotID, repoPath, restorePaths, hasHDD)
// Stop the app before restore
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: step 1/4 — stopping app %s", stackName)
}
if err := m.stackProvider.StopStack(stackName); err != nil {
m.logger.Printf("[WARN] RESTORE could not stop %s: %v (proceeding anyway)", stackName, err)
}
// Execute restore via restic
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: step 2/4 — restoring data from snapshot %s", snapshotID)
}
if err := m.restic.RestoreAppData(repoPath, snapshotID, restorePaths); err != nil {
m.logger.Printf("[ERROR] RESTORE failed for %s: %v", stackName, err)
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: step 3/4 — restarting app %s after failure", stackName)
}
if startErr := m.stackProvider.StartStack(stackName); startErr != nil {
m.logger.Printf("[WARN] RESTORE could not restart %s after failure: %v", stackName, startErr)
}
return err
}
// Restart the app
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: step 3/4 — restarting app %s after successful restore", stackName)
}
if err := m.stackProvider.StartStack(stackName); err != nil {
m.logger.Printf("[WARN] RESTORE could not restart %s after restore: %v", stackName, err)
}
restoreType := "config+DB"
if hasHDD {
restoreType = "full (config+DB+userdata)"
}
if m.isDebug() {
m.logger.Printf("[DEBUG] RestoreApp: step 4/4 — restore completed, type=%s", restoreType)
}
m.logger.Printf("[INFO] RESTORE completed: stack=%s, snapshot=%s, type=%s", stackName, snapshotID, restoreType)
return nil
}