v0.142.0: offsite repo continuity — orphaned-repo guard (A) + run-status auto-refresh (C)

- Part A: classify restic cat-config failure (wrong-password=orphaned vs no-repo vs other); ORPHANED state + Hungarian card + offbox_repo_orphaned/reset events (once, not nightly); reset = move-aside (never delete) + init, unclaimed auto / claimed confirm. Red-proofs TestOffbox_OrphanDetection_* + ConfirmedReset.
- Part C: GET /backup/offbox/status + poll on backups_remote → flips Fut→Rendben/Hiba without manual reload.
This commit is contained in:
2026-07-17 10:47:30 +02:00
parent 1452dd2b17
commit 596505ed64
11 changed files with 587 additions and 18 deletions
+215 -17
View File
@@ -6,6 +6,7 @@ import (
"crypto/sha256"
"encoding/hex"
"encoding/json"
"errors"
"fmt"
"os"
"os/exec"
@@ -53,6 +54,162 @@ func (m *Manager) SetOffboxNotify(fn func(dur time.Duration, snapshots int, err
m.offboxNotify = fn
}
// SetOffboxOrphanEvent wires the offsite-repo continuity event push (main.go → notifier).
func (m *Manager) SetOffboxOrphanEvent(fn func(eventType, renamedTo string)) { m.offboxOrphanEvent = fn }
// SetOffboxSSH overrides the raw-ssh exec used for the orphaned-repo move-aside (tests).
func (m *Manager) SetOffboxSSH(fn func(ctx context.Context, host, user string, port int, keyPath, knownHosts, remoteCmd string) ([]byte, error)) {
m.offboxSSH = fn
}
// ErrOffboxOrphaned is the sentinel returned when the offsite repo exists but is keyed under a
// passphrase this controller no longer has (the reinstall shape) — the run skips and the UI shows
// the orphan card instead of the raw restic error.
var ErrOffboxOrphaned = fmt.Errorf("offbox repo orphaned: exists but keyed under a previous, no-longer-available passphrase")
// classifyResticProbe maps a `restic cat config` failure to a repo class. The signatures are the exact
// restic stderr matched in the 2026-07-17 diagnosis + restic's no-repo message:
// - "orphaned": repo present, wrong key ("wrong password or no key found") — the definitive signal
// - "norepo": no repo at the location (init is the correct path)
// - "other": network/SFTP-auth/unknown — NOT orphaned; existing error handling
func classifyResticProbe(out []byte, err error) string {
if err == nil {
return "" // success — repo good
}
s := strings.ToLower(string(out))
switch {
case strings.Contains(s, "wrong password or no key found"):
return "orphaned"
case strings.Contains(s, "unable to open config file"),
strings.Contains(s, "is there a repository at the following location"),
strings.Contains(s, "no such file"),
strings.Contains(s, "does not exist"):
return "norepo"
default:
return "other"
}
}
func defaultOffboxSSH(ctx context.Context, host, user string, port int, keyPath, knownHosts, remoteCmd string) ([]byte, error) {
if port == 0 {
port = 22
}
args := []string{
"-p", fmt.Sprint(port), "-oBatchMode=yes", fmt.Sprintf("-oConnectTimeout=%d", offboxConnectTimeoutSec),
"-oStrictHostKeyChecking=yes", "-oUserKnownHostsFile=" + knownHosts, "-i", keyPath,
user + "@" + host, remoteCmd,
}
cmd := exec.CommandContext(ctx, "ssh", args...)
return cmd.CombinedOutput()
}
func (m *Manager) sshRunner() func(ctx context.Context, host, user string, port int, keyPath, knownHosts, remoteCmd string) ([]byte, error) {
if m.offboxSSH != nil {
return m.offboxSSH
}
return defaultOffboxSSH
}
// OffboxOrphaned reports whether the offsite repo is in the ORPHANED state (persisted).
func (m *Manager) OffboxOrphaned() bool {
t := m.settings.GetOffboxTarget()
return t != nil && t.RepoState == "orphaned"
}
// OffboxOrphanedRenamedTo returns the last move-aside path (for the card copy; "" if none).
func (m *Manager) OffboxOrphanedRenamedTo() string {
t := m.settings.GetOffboxTarget()
if t == nil {
return ""
}
return t.OrphanedRenamedTo
}
// markOrphaned sets the persistent ORPHANED state and, ONLY on the transition into it (not already
// orphaned), pushes the offbox_repo_orphaned event — so scheduled runs never nightly-spam.
func (m *Manager) markOrphaned() {
already := m.OffboxOrphaned()
if err := m.settings.UpdateOffboxStatus(func(o *settings.OffboxTarget) {
o.RepoState = "orphaned"
if o.OrphanedAt == "" || !already {
o.OrphanedAt = time.Now().UTC().Format(time.RFC3339)
}
}); err != nil {
m.logger.Printf("[WARN] [offbox] persist orphaned state failed: %v", err)
}
if !already {
m.logger.Printf("[WARN] [offbox] offsite repo ORPHANED — remote holds backups written under a previous, no-longer-available key; runs will skip until reset")
if m.offboxOrphanEvent != nil {
m.offboxOrphanEvent("offbox_repo_orphaned", "")
}
}
}
// resetOrphanedRepo moves the orphaned repo aside (never deletes) and re-inits a fresh repo under the
// CURRENT passphrase. Reversible. Used by the unclaimed auto-reset (Scenario B) and the claimed
// confirmed reset (Scenario C). Caller holds the single-flight guarantee (run mutex) OR is the handler.
func (m *Manager) resetOrphanedRepo(ctx context.Context, base, env []string, reason string) error {
t := m.settings.GetOffboxTarget()
if t == nil {
return fmt.Errorf("no offsite target configured")
}
port := t.Port
if port == 0 {
port = 22
}
// Choose a move-aside name that never overwrites an earlier orphaned copy (edge rule: -2, -3).
date := time.Now().UTC().Format("20060102")
base1 := t.RepoPath + ".orphaned-" + date
newPath := base1
for i := 2; i <= 20; i++ {
// `test -e <p>` returns non-zero (exit 1) when absent — that is the name we want. A transport
// error also lands here; we then just try the mv and let it fail loudly rather than loop.
out, err := m.sshRunner()(ctx, t.Host, t.User, port, m.offboxKeyPath(), m.offboxKnownHosts(), "test -e "+shellQuote(newPath))
if err != nil && !strings.Contains(strings.ToLower(string(out)), "denied") {
break // absent (test -e exit 1) → free name
}
newPath = fmt.Sprintf("%s-%d", base1, i)
}
m.logger.Printf("[WARN] [offbox] resetting orphaned repo (%s): move-aside %s -> %s, then re-init", reason, t.RepoPath, newPath)
if out, err := m.sshRunner()(ctx, t.Host, t.User, port, m.offboxKeyPath(), m.offboxKnownHosts(),
fmt.Sprintf("mv %s %s", shellQuote(t.RepoPath), shellQuote(newPath))); err != nil {
return fmt.Errorf("offbox move-aside failed: %w: %s", err, truncate(out))
}
// Fresh init under the current passphrase.
ictx, icancel := context.WithTimeout(ctx, offboxProbeTimeout)
defer icancel()
if out, err := m.runner()(ictx, env, append(append([]string{}, base...), "init")...); err != nil {
return fmt.Errorf("offbox re-init after move-aside failed: %w: %s", err, truncate(out))
}
if err := m.settings.UpdateOffboxStatus(func(o *settings.OffboxTarget) {
o.RepoState = ""
o.OrphanedAt = ""
o.OrphanedRenamedTo = newPath
o.LastError = ""
}); err != nil {
m.logger.Printf("[WARN] [offbox] clear orphaned state failed: %v", err)
}
m.logger.Printf("[INFO] [offbox] orphaned repo reset complete — old history set aside at %s (move-aside, not deleted); fresh repo initialized", newPath)
if m.offboxOrphanEvent != nil {
m.offboxOrphanEvent("offbox_repo_reset", newPath)
}
return nil
}
// ResetOrphanedRepo is the handler entry point for the CLAIMED confirmed reset (Scenario C). It refuses
// unless the repo is currently orphaned. It builds the base/env and runs the move-aside + re-init.
func (m *Manager) ResetOrphanedRepo(ctx context.Context) error {
if !m.OffboxOrphaned() {
return fmt.Errorf("az offsite tároló nincs elárvult állapotban")
}
t := m.settings.GetOffboxTarget()
base, env := m.offboxBaseArgs(t)
return m.resetOrphanedRepo(ctx, base, env, "operator-confirmed (claimed)")
}
// shellQuote single-quotes a path for the remote shell (our repo paths have no single quotes).
func shellQuote(s string) string { return "'" + strings.ReplaceAll(s, "'", `'\''`) + "'" }
// SetOffboxSizer overrides the mandatory-set byte estimator (tests). SetOffboxEnlargeBlockedNotifier
// wires the edge-triggered enlargement-blocked notification (main.go). SetOffboxPlaceCopier overrides
// the place-to-live missing-only merge (tests).
@@ -368,23 +525,47 @@ func (m *Manager) resticStep(ctx context.Context, env, base []string, label stri
func (m *Manager) ensureOffboxRepo(ctx context.Context, base, env []string) error {
pctx, cancel := context.WithTimeout(ctx, offboxProbeTimeout)
defer cancel()
if _, err := m.runner()(pctx, env, append(append([]string{}, base...), "cat", "config")...); err == nil {
return nil // repo exists
}
// Repo (probably) absent OR unreachable. Try init; if init succeeds the repo was absent. If init
// fails because it already exists (a race), treat as success; otherwise the error is real (e.g. a
// dead NAS — fail fast).
ictx, icancel := context.WithTimeout(ctx, offboxProbeTimeout)
defer icancel()
out, err := m.runner()(ictx, env, append(append([]string{}, base...), "init")...)
if err == nil {
m.logger.Printf("[INFO] [offbox] initialized restic repo")
pout, perr := m.runner()(pctx, env, append(append([]string{}, base...), "cat", "config")...)
switch classifyResticProbe(pout, perr) {
case "": // success — repo good (and clear any stale orphaned flag)
if m.OffboxOrphaned() {
_ = m.settings.UpdateOffboxStatus(func(o *settings.OffboxTarget) { o.RepoState = ""; o.OrphanedAt = "" })
}
return nil
case "orphaned":
// The repo EXISTS but is keyed under a passphrase we no longer have (the reinstall shape). An
// UNCLAIMED (as-delivered) box auto-resets (Scenario B); a CLAIMED box surfaces the orphan card
// and skips until the customer confirms a reset (Scenario C). Move-aside, never delete.
if !m.settings.GetClaimed() {
m.logger.Printf("[INFO] [offbox] orphaned repo on an UNCLAIMED box — auto-resetting (move-aside + re-init)")
if m.offboxOrphanEvent != nil {
m.offboxOrphanEvent("offbox_repo_orphaned", "")
}
if rerr := m.resetOrphanedRepo(ctx, base, env, "auto (unclaimed)"); rerr != nil {
m.markOrphaned() // auto-reset failed → fall back to the orphan card so it isn't silent
return ErrOffboxOrphaned
}
return nil // repo is fresh under the current passphrase → the run proceeds
}
m.markOrphaned()
return ErrOffboxOrphaned
case "norepo":
// No repo at the location → init (the normal first-run path). A race where it already exists is
// treated as success; any other init error is real (e.g. a dead NAS — fail fast).
ictx, icancel := context.WithTimeout(ctx, offboxProbeTimeout)
defer icancel()
out, err := m.runner()(ictx, env, append(append([]string{}, base...), "init")...)
if err == nil {
m.logger.Printf("[INFO] [offbox] initialized restic repo")
return nil
}
if strings.Contains(string(out), "already initialized") || strings.Contains(string(out), "already exists") {
return nil
}
return fmt.Errorf("offbox repo unreachable / init failed: %w: %s", err, truncate(out))
default: // "other" — network/SFTP-auth/unknown; NOT orphaned. Surface as before (fail fast).
return fmt.Errorf("offbox repo unreachable: %w: %s", perr, truncate(pout))
}
if strings.Contains(string(out), "already initialized") || strings.Contains(string(out), "already exists") {
return nil
}
return fmt.Errorf("offbox repo unreachable / init failed: %w: %s", err, truncate(out))
}
// RunOffboxBackup backs up every off-box-toggled app's recovery unit (recovery unit + DB dumps + volume
@@ -400,6 +581,12 @@ func (m *Manager) RunOffboxBackup(ctx context.Context) error {
m.logger.Printf("[INFO] [offbox] skipped — pending key escrow (no offsite run until the repo password is escrowed under R)")
return nil
}
// Offsite-repo continuity (v0.142.0): once ORPHANED, scheduled runs SKIP (the event fired on the
// detection transition — no nightly spam) until a reset clears it. The remote page shows the card.
if m.OffboxOrphaned() {
m.logger.Printf("[INFO] [offbox] skipped — offsite repo orphaned (awaiting reset)")
return nil
}
if m.migrationActive() {
m.logger.Printf("[INFO] [offbox] skipped — migration in progress")
return nil
@@ -467,7 +654,13 @@ func (m *Manager) RunOffboxBackup(ctx context.Context) error {
if perr := m.settings.UpdateOffboxStatus(func(o *settings.OffboxTarget) {
o.LastRun = time.Now().UTC().Format(time.RFC3339)
o.LastDuration = dur.Round(time.Second).String()
if runErr != nil {
if errors.Is(runErr, ErrOffboxOrphaned) {
// First-detection of the orphaned repo: RepoState (set by markOrphaned) drives the orphan
// card — do NOT surface the raw restic/sentinel text as the last-error banner.
o.LastStatus = "error"
o.LastError = ""
o.LastWarning = ""
} else if runErr != nil {
o.LastStatus = "error"
o.LastError = runErr.Error()
o.LastWarning = ""
@@ -503,7 +696,9 @@ func (m *Manager) RunOffboxBackup(ctx context.Context) error {
}); perr != nil {
m.logger.Printf("[WARN] [offbox] status persist (final) failed: %v", perr)
}
if m.offboxNotify != nil {
// The orphaned case has its OWN dedicated event (offbox_repo_orphaned) — do NOT also fire the
// generic backup-failed notification (no double/raw alert; the orphan card is the customer surface).
if m.offboxNotify != nil && !errors.Is(runErr, ErrOffboxOrphaned) {
m.offboxNotify(dur, snapshots, runErr)
}
// Edge-triggered enlarge-blocked notification: only apps that NEWLY crossed into the blocked state
@@ -518,6 +713,9 @@ func (m *Manager) RunOffboxBackup(ctx context.Context) error {
}
}
switch {
case errors.Is(runErr, ErrOffboxOrphaned):
m.logger.Printf("[WARN] [offbox] run skipped — offsite repo orphaned (card shown; awaiting reset)")
return nil // the orphaned STATE + event are the signal; not a hard run error for the scheduler
case runErr != nil:
m.logger.Printf("[ERROR] [offbox] backup failed after %s: %v", dur.Round(time.Second), runErr)
case len(missing) > 0: