felhom-controller/controller/internal/backup/restore_unit.go

package backup

import (
	"fmt"
	"os"
	"path/filepath"
	"strings"
	"time"

	"gopkg.in/yaml.v3"
)

// reconcileRestoreSecrets merges the recovery unit's non-secret env with the secrets recovered from
// the unit itself (D5) and from the guest's own app.yaml, and applies the FAIL-CLOSED data-key gate.
// It is the safety-critical heart of Phase 2b and is deliberately a pure function (no I/O) so it can
// be exhaustively unit-tested — the D5 source arrives as an ARGUMENT, not as a read.
//
// Policy:
//   - Regenerate NOTHING here. Secrets come from the unit (portable class) or the guest (the rest).
//   - A missing DATA-ENCRYPTING key (`dataKeyNames`) is FATAL: regenerating it would render the
//     restored data unreadable, so we refuse and tell the operator to do a PBS whole-guest restore.
//     D5 means the key is normally IN the unit — but "normally" is not a reason to soften the gate.
//   - A missing resettable secret is NON-fatal: returned in `missing` so the caller can warn or
//     regenerate it (O4). No data is lost.
//
// PRECEDENCE — the UNIT WINS over the guest when both hold a value for the same name.
//
// This is not arbitrary and it is not "newest wins". The unit's secrets are captured in the SAME run
// as the dumps beside them (runVolumeDumps → captureAllRecoveryUnits, backup.go), so the unit's value
// is the one that MATCHES THE DATA ABOUT TO BE RESTORED, whereas the guest's value is merely the most
// recent. Where they disagree the guest's has been rotated since the capture, and preferring it is
// precisely the data-loss bug:
//   - a rotated data-encrypting key does not decrypt data encrypted with the old one;
//   - a rotated DB password does not match the scram/mysql hash inside the restored data directory
//     (POSTGRES_PASSWORD is ignored once PGDATA is non-empty), so the app cannot reach its own rows.
//
// The restore persists fullEnv back to the guest's app.yaml (RecreateStackDefinitionFromUnit), so
// unit-wins also leaves the guest consistent with the data now on disk.
func reconcileRestoreSecrets(nonSecretEnv, unitSecrets, guestSecrets map[string]string, secretNames, dataKeyNames []string) (fullEnv map[string]string, missing []string, err error) {
	fullEnv = make(map[string]string, len(nonSecretEnv)+len(secretNames))
	for k, v := range nonSecretEnv {
		fullEnv[k] = v
	}
	// resolve applies the precedence: unit first, guest only as a fallback.
	resolve := func(n string) (string, bool) {
		if v, ok := unitSecrets[n]; ok && v != "" {
			return v, true
		}
		if v, ok := guestSecrets[n]; ok && v != "" {
			return v, true
		}
		return "", false
	}
	have := func(n string) bool {
		_, ok := resolve(n)
		return ok
	}
	for _, n := range secretNames {
		if v, ok := resolve(n); ok {
			fullEnv[n] = v
		} else {
			missing = append(missing, n)
		}
	}
	// Fail-closed: any unrecoverable data-encrypting key aborts the restore.
	var missingDataKeys []string
	for _, dk := range dataKeyNames {
		if !have(dk) {
			missingDataKeys = append(missingDataKeys, dk)
		}
	}
	if len(missingDataKeys) > 0 {
		return nil, missing, fmt.Errorf(
			"refusing to restore: data-encrypting key(s) %v are in NEITHER the recovery unit nor the guest's app.yaml — "+
				"a PBS whole-guest restore is required first (regenerating the key would render stored data unreadable)",
			missingDataKeys)
	}
	return fullEnv, missing, nil
}

// readUnitEnv parses a recovery unit's app.yaml and SPLITS it into the plain config env and the
// secrets the unit carries (D5), using the manifest's portable-secret names as the discriminator.
//
// The split is driven by the MANIFEST, not by guessing from key names: the manifest and the app.yaml
// are captured together and checksummed together, so they cannot disagree about which entries are
// secrets. A schema-1 unit has no portable names, so everything lands in nonSecret — exactly the
// pre-D5 behaviour, which is what makes an old unit still restorable.
func readUnitEnv(path string, portableNames []string) (nonSecret, unitSecrets map[string]string) {
	nonSecret, unitSecrets = map[string]string{}, map[string]string{}
	data, err := os.ReadFile(path)
	if err != nil {
		return nonSecret, unitSecrets
	}
	var s strippedAppYaml
	if yaml.Unmarshal(data, &s) != nil || s.Env == nil {
		return nonSecret, unitSecrets
	}
	isPortable := make(map[string]bool, len(portableNames))
	for _, n := range portableNames {
		isPortable[n] = true
	}
	for k, v := range s.Env {
		if isPortable[k] {
			unitSecrets[k] = v
			continue
		}
		nonSecret[k] = v
	}
	return nonSecret, unitSecrets
}

// hasReplayableDump reports whether dumpDir holds a .sql dump that the replay could actually use.
// The `pre-restore-` safety dumps are EXCLUDED: they live in the same directory (deliberately — an
// undo the customer cannot see is not much of one) but are never a replay source, so counting them
// would arm the DB-only phase, and its fail-closed gate, for an app that has nothing to replay.
func hasReplayableDump(dumpDir string) bool {
	entries, err := os.ReadDir(dumpDir)
	if err != nil {
		return false
	}
	for _, e := range entries {
		if e.IsDir() || filepath.Ext(e.Name()) != ".sql" {
			continue
		}
		if !strings.HasPrefix(e.Name(), preRestoreDumpPrefix) {
			return true
		}
	}
	return false
}

// UnitRestoreResult is what a local recovery-unit restore actually did, so the surface can STATE it
// rather than report a bare completion.
//
// It exists for the same reason OffsiteReconstituteResult does, and it is the same lesson arriving on
// the other path: on 2026-08-21 an opengist restore reported "Restore-from-unit completed" over a unit
// holding manifest.json and compose/ and nothing else, and no screen could have told the customer that
// no data had been returned (R-353).
//
// The Manifest* counts are carried BECAUSE zero-replayed has two causes and they are not the same
// fact. A unit that lists no dumps means THE BACKUP held no data. A unit that lists dumps none of which
// replayed means something is wrong and the customer's live data was left untouched. R-355 is the
// standing rule this obeys: a claim about the APP must never be inferred from a counter — and here it
// is not merely unproven but unprovable, because 07-backup-architecture §6.3 records that an app's
// canonical .sql could be absent from the unit for reasons that have nothing to do with whether the app
// has a database (R-361 destroyed exactly that file for four months).
type UnitRestoreResult struct {
	// VolumesReplayed is how many named-volume tars were unpacked into live Docker volumes.
	VolumesReplayed int
	// DBsReplayed is how many .sql dumps were imported. Never inferred from the presence of a database
	// service — only a completed import increments it.
	DBsReplayed int
	// ManifestVolumes is len(manifest.VolumeDumps): what the unit CLAIMS it captured. The gap between
	// this and VolumesReplayed is the whole of Scenario C.
	ManifestVolumes int
	// ManifestDBs is len(manifest.DBDumps): the same claim for the database leg.
	ManifestDBs int
	// CountsUnknown marks a run whose counts could not be established AT ALL — today the one case is
	// the no-unit fallback to RestoreApp, which returns only an error and whose signature is
	// deliberately out of scope.
	//
	// IT EXISTS BECAUSE THE ZERO VALUE WOULD OTHERWISE LIE. Without it a fallback restore that really
	// replayed three volumes reports VolumesReplayed=0 / ManifestVolumes=0 — the shape the surface
	// reads as „ez a mentés csak a beállításokat tartalmazta, adatot nem". That is a confident false
	// statement, and precisely the R-88 failure direction (degrade to NO DATA rather than to UNKNOWN)
	// this whole change exists to remove. An unknown must be carried, never drawn as a zero.
	CountsUnknown bool
}

// RestoreFromRecoveryUnit recreates an app from its on-drive recovery unit.
//
// It reads the unit manifest, takes the portable secrets from the UNIT and the rest from the guest's
// live app.yaml (unit wins — see reconcileRestoreSecrets), applies the fail-closed data-key gate,
// restores the named-volume data from the unit's tars, then restores the app's definition from the unit
// and redeploys it with the reconstructed env (re-pulling the pinned image). If no unit exists it falls
// back to the legacy volume-only RestoreApp.
//
// D5: this no longer needs the guest. A restore with the guest's app.yaml absent succeeds, which is
// pinned by TestRestoreFromRecoveryUnitWithGuestAbsent — the withheld class is regenerated (O4) and
// only a data key missing from BOTH sources still refuses.
func (m *Manager) RestoreFromRecoveryUnit(stackName string) (UnitRestoreResult, error) {
	var res UnitRestoreResult
	if m.stackProvider == nil {
		return res, fmt.Errorf("stack provider not configured")
	}

	m.mu.Lock()
	if m.running {
		m.mu.Unlock()
		return res, 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()
	}()

	drivePath := m.GetAppDrivePath(stackName)
	if drivePath == "" || !filepath.IsAbs(drivePath) {
		return res, fmt.Errorf("cannot determine drive path for %s", stackName)
	}
	nsRoot := m.namespaceRoot(drivePath)

	manifest := readManifest(RecoveryUnitManifestPath(nsRoot, stackName))
	if manifest == nil {
		m.logger.Printf("[WARN] [backup] No recovery unit for %s — falling back to volume-only restore", stackName)
		m.mu.Lock()
		m.running = false // RestoreApp re-acquires the running flag
		m.mu.Unlock()
		// The fallback has no unit and no manifest, and `RestoreApp` returns only an error — so the
		// counts here are genuinely UNKNOWN, not zero. Reporting zero would tell a customer whose
		// volumes were just restored that their backup held no data.
		res.CountsUnknown = true
		return res, m.RestoreApp(stackName, "")
	}

	// R-353: what the unit CLAIMS it holds, recorded before any mutation. Read from the manifest that
	// was just parsed above, so the claim and the outcome are counted from the same document.
	res.ManifestVolumes, res.ManifestDBs = len(manifest.VolumeDumps), len(manifest.DBDumps)

	composeDir := RecoveryUnitComposePath(nsRoot, stackName)
	nonSecretEnv, unitSecrets := readUnitEnv(filepath.Join(composeDir, "app.yaml"), manifest.PortableSecretEnvVars)

	// D5: the unit carries the portable class, so this is the leg that no longer needs the guest. The
	// guest is still consulted for the WITHHELD class (internet-reachable admin logins) and as the
	// fallback for a schema-1 unit — it returns an empty map when the guest is gone, which is the whole
	// point: a Tier-1/2 restore must survive that. Precedence is unit-over-guest (see
	// reconcileRestoreSecrets), then the fail-closed gate.
	guestSecrets := m.stackProvider.RecoverStackSecrets(stackName, manifest.SecretEnvVars)
	fullEnv, missing, err := reconcileRestoreSecrets(nonSecretEnv, unitSecrets, guestSecrets, manifest.SecretEnvVars, manifest.DataKeyEnvVars)
	if err != nil {
		m.logger.Printf("[ERROR] [backup] Restore REFUSED for %s: %v", stackName, err)
		return res, err
	}
	// O4: a missing RESETTABLE secret used to redeploy blank (compose "Defaulting to a blank
	// string" → exit 1). Generate a replacement via the deploy flow's generator instead —
	// RecreateStackFromUnit persists fullEnv through SaveAppConfig, so the new value lands
	// encrypted in the guest app.yaml and round-trips on the next backup/restore. Data-keys are
	// never generated: the fail-closed gate above already refused if one was missing, and the
	// generator itself refuses data-key fields (defense-in-depth). Values are never logged.
	//
	// D5 shrinks this path to the rare case: the portable class now comes from the unit, so a
	// generator run means the secret was empty at capture AND absent from the guest.
	//
	// It does NOT claim the reset is harmless. R-127: for a DB password it is not — a restored data
	// directory keeps the OLD role hash (POSTGRES_PASSWORD is ignored once PGDATA is non-empty), so a
	// regenerated value leaves the app unable to authenticate against its own restored rows while the
	// dump replay, which uses the container's local trust socket, still reports success. The old wording
	// here asserted "stored data is unaffected" for every non-data-key secret; that is false for the 18
	// DB/root-password fields and is now scoped to what is actually true.
	if len(missing) > 0 {
		dataKeySet := make(map[string]bool, len(manifest.DataKeyEnvVars))
		for _, dk := range manifest.DataKeyEnvVars {
			dataKeySet[dk] = true
		}
		var generated, unresolved []string
		for _, name := range missing {
			if !dataKeySet[name] && m.generateSecret != nil {
				if v, ok := m.generateSecret(stackName, name); ok && v != "" {
					fullEnv[name] = v
					generated = append(generated, name)
					continue
				}
			}
			unresolved = append(unresolved, name)
		}
		if len(generated) > 0 {
			m.logger.Printf("[WARN] [backup] Restore %s: generated replacement for %v — the credential was reset (old value unrecoverable); no data-encrypting key was involved, but a regenerated DATABASE password will not match the restored data directory's stored hash (R-127) — check the app can reach its data",
				stackName, generated)
		}
		if len(unresolved) > 0 {
			m.logger.Printf("[WARN] [backup] Restore %s: %d resettable secret(s) unrecoverable and have no generator %v — proceeding, but the app may fail to start until the credential is set manually",
				stackName, len(unresolved), unresolved)
		}
	}
	m.logger.Printf("[INFO] [backup] Restoring %s from recovery unit: images=%d, secrets recovered=%d/%d, data_keys=%d",
		stackName, len(manifest.ImagePins), len(manifest.SecretEnvVars)-len(missing), len(manifest.SecretEnvVars), len(manifest.DataKeyEnvVars))

	// R-47: which compose service holds the database, and is there anything to replay? Resolved from
	// the UNIT's compose, because that file is about to BECOME the live one. Both answers are needed
	// BEFORE the first mutation, so the refusal below leaves the live app completely untouched.
	dbServices, dsErr := DBServiceNames(filepath.Join(composeDir, "docker-compose.yml"))
	if dsErr != nil {
		// "cannot tell" is not "no database" — leave it empty and let the gate decide.
		m.logger.Printf("[WARN] [backup] %s: could not read the unit's compose services: %v", stackName, dsErr)
	}
	hasDumps := hasReplayableDump(AppDBDumpPath(nsRoot, stackName))
	if hasDumps && len(dbServices) == 0 {
		m.logger.Printf("[ERROR] [backup] Restore REFUSED for %s: a .sql dump exists but no database service is identifiable in the unit's compose", stackName)
		return res, fmt.Errorf("Az adatbázis-szolgáltatás nem azonosítható a(z) %s alkalmazásban — a visszaállítás biztonsági okból nem indult el.", stackName)
	}

	// Stop, restore named-volume data, recreate the definition, replay the DB with ONLY the database
	// service running, and only then start the whole stack.
	// F17: surface a data-restore failure instead of swallowing it (we still bring the app back up).
	var dataErr error
	if err := m.stackProvider.StopStack(stackName); err != nil {
		m.logger.Printf("[WARN] [backup] could not stop %s before restore: %v (continuing)", stackName, err)
	}
	// R-353: the count is captured even when the replay errors — a partial replay is a fact the
	// customer's sentence has to be built from, and discarding it on the error path is how Scenario C
	// would end up wearing Scenario B's wording.
	replayed, volErr := m.restoreDockerVolumes(stackName, drivePath)
	res.VolumesReplayed = replayed
	if volErr != nil {
		m.logger.Printf("[ERROR] [backup] volume restore for %s: %v", stackName, volErr)
		dataErr = volErr
	}
	if err := m.stackProvider.RecreateStackDefinitionFromUnit(stackName, composeDir, fullEnv); err != nil {
		return res, fmt.Errorf("recreating %s from unit: %w", stackName, err)
	}
	// F17: the captured .sql dump is the authoritative logical DB state — replay it AFTER the volume
	// restore, so the dump WINS over any volume-tar copy of the database.
	// R-47: the replay happens with ONLY the database service up. This used to run after
	// RecreateStackFromUnit had already brought the WHOLE stack up, letting the application rebuild
	// schema objects underneath the replay (H4, DIAG-immich-restore-round2-2026-07-19).
	if hasDumps {
		if err := m.stackProvider.StartStackServices(stackName, dbServices); err != nil {
			m.logger.Printf("[ERROR] [backup] DB-only start for %s: %v", stackName, err)
			if dataErr == nil {
				dataErr = err
			}
		} else if n, err := m.reimportDBDumpsCtx(stackName, nsRoot); err != nil {
			res.DBsReplayed = n // partial credit: whatever imported before the failure really did import
			m.logger.Printf("[ERROR] [backup] DB re-import for %s: %v", stackName, err)
			if dataErr == nil {
				dataErr = err
			}
		} else {
			res.DBsReplayed = n
		}
	}
	if err := m.stackProvider.StartStack(stackName); err != nil {
		return res, fmt.Errorf("starting %s after restore from unit: %w", stackName, err)
	}
	if err := m.waitForHealthy(stackName, 90*time.Second); err != nil {
		m.logger.Printf("[WARN] [backup] %s restored but health check failed: %v", stackName, err)
	}

	if dataErr != nil {
		return res, fmt.Errorf("restore of %s from unit completed with data errors: %w", stackName, dataErr)
	}
	m.logger.Printf("[INFO] [backup] Restore-from-unit completed: %s — %d volume(s) of %d listed, %d database(s) of %d listed",
		stackName, res.VolumesReplayed, res.ManifestVolumes, res.DBsReplayed, res.ManifestDBs)
	return res, nil
}