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:
@@ -588,6 +588,18 @@ func main() {
|
||||
"A(z) %s teljes távoli mentése (~%s) túllépné a tárhelykeretet (%d/%d GB). A konfiguráció és az adatbázis továbbra is mentésre kerül; nagyobb kerethez vedd fel velünk a kapcsolatot.",
|
||||
stack, appbackup.HumanizeBytes(estBytes), usedGB, quotaGB))
|
||||
})
|
||||
// v0.142.0 offsite-repo continuity: push a hub event on the orphaned/reset transitions (once per
|
||||
// transition — the Manager guards nightly re-fire). Operator-visible on the customer page.
|
||||
backupMgr.SetOffboxOrphanEvent(func(eventType, renamedTo string) {
|
||||
switch eventType {
|
||||
case "offbox_repo_orphaned":
|
||||
notifier.PushEvent("offbox_repo_orphaned", "warning",
|
||||
"A távoli mentési tároló elárvult: a benne lévő mentések egy korábbi, már nem elérhető kulccsal készültek (újratelepítés). Új mentés a tároló visszaállításáig nem készül.", nil)
|
||||
case "offbox_repo_reset":
|
||||
notifier.PushEvent("offbox_repo_reset", "info",
|
||||
"A távoli mentési tároló visszaállítva: a régi előzmény félretéve (nem törölve), és egy üres, új tároló jött létre a mostani kulccsal.", map[string]string{"renamed_to": renamedTo})
|
||||
}
|
||||
})
|
||||
sched.Daily("offbox-backup", "04:15", func(ctx context.Context) error {
|
||||
t := sett.GetOffboxTarget()
|
||||
if t == nil || !t.Enabled || t.Schedule != "daily" || !backupMgr.OffboxConfigured() {
|
||||
|
||||
@@ -35,6 +35,13 @@ type Manager struct {
|
||||
// offbox (Part B): the restic-SFTP exec seam (nil → real restic) + the failure→operator-alert hook.
|
||||
offboxRunner offboxRunner
|
||||
offboxNotify func(dur time.Duration, snapshots int, err error)
|
||||
// offboxOrphanEvent (v0.142.0), if set, pushes a hub event on offsite-repo continuity transitions
|
||||
// ("offbox_repo_orphaned" / "offbox_repo_reset"); renamedTo names the move-aside path (reset only).
|
||||
// Wired in main.go to the notifier. Nil-safe.
|
||||
offboxOrphanEvent func(eventType, renamedTo string)
|
||||
// offboxSSH (v0.142.0) is the raw-ssh exec seam for the orphaned-repo move-aside (restic has no
|
||||
// rename); tests inject a fake. Nil → the real ssh invocation (defaultOffboxSSH).
|
||||
offboxSSH func(ctx context.Context, host, user string, port int, keyPath, knownHosts, remoteCmd string) ([]byte, error)
|
||||
|
||||
// offboxSizer (3a) — the mandatory-set byte estimator for the pre-push enlargement gate, overridable
|
||||
// in tests so the gate is unit-testable without a real du. Nil → the real dirSizeBytes (du -sb).
|
||||
|
||||
@@ -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:
|
||||
|
||||
@@ -0,0 +1,172 @@
|
||||
package backup
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"strings"
|
||||
"testing"
|
||||
)
|
||||
|
||||
// classifyResticProbe maps the exact restic stderr to a repo class (the 2026-07-17 diagnosis
|
||||
// signatures). ORPHANED only on the definitive wrong-password line; ambiguous errors are NOT orphaned.
|
||||
func TestClassifyResticProbe(t *testing.T) {
|
||||
cases := []struct {
|
||||
out string
|
||||
err error
|
||||
want string
|
||||
}{
|
||||
{"", nil, ""}, // success
|
||||
{"Fatal: wrong password or no key found", fmt.Errorf("exit status 1"), "orphaned"},
|
||||
{"Fatal: unable to open config file: <sftp:...> does not exist\nIs there a repository at the following location?", fmt.Errorf("exit status 1"), "norepo"},
|
||||
{"ssh: connect to host nas.local port 22: Connection timed out", fmt.Errorf("exit status 255"), "other"},
|
||||
{"Load(<lock/...>): permission denied", fmt.Errorf("exit status 1"), "other"},
|
||||
}
|
||||
for _, c := range cases {
|
||||
if got := classifyResticProbe([]byte(c.out), c.err); got != c.want {
|
||||
t.Errorf("classify(%q) = %q, want %q", c.out, got, c.want)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// wrongPwRunner: `cat config` returns the wrong-password signature; other restic steps succeed (so a
|
||||
// post-reset run can proceed). Records the subcommands seen.
|
||||
func wrongPwRunner(seen *[]string) offboxRunner {
|
||||
return func(_ context.Context, _ []string, args ...string) ([]byte, error) {
|
||||
sub := ""
|
||||
for i, a := range args {
|
||||
if a == "cat" && i+1 < len(args) && args[i+1] == "config" {
|
||||
sub = "cat-config"
|
||||
} else if a == "init" {
|
||||
sub = "init"
|
||||
}
|
||||
}
|
||||
if sub == "" && len(args) > 0 {
|
||||
sub = args[len(args)-1]
|
||||
}
|
||||
if seen != nil {
|
||||
*seen = append(*seen, sub)
|
||||
}
|
||||
if sub == "cat-config" {
|
||||
return []byte("Fatal: wrong password or no key found"), fmt.Errorf("exit status 1")
|
||||
}
|
||||
return nil, nil // init / unlock / backup / stats succeed
|
||||
}
|
||||
}
|
||||
|
||||
// Scenario A (RED-PROOF = the incident): a CLAIMED box whose repo is wrong-keyed enters the explicit
|
||||
// ORPHANED state — the run skips cleanly (no raw restic banner, ONE event, no nightly re-fire) instead
|
||||
// of erroring nightly with "exit status 1". Pre-fix (no classification) surfaced the raw error and set
|
||||
// no state → these assertions FAIL.
|
||||
func TestOffbox_OrphanDetection_Claimed(t *testing.T) {
|
||||
m, sett := newOffboxManager(t)
|
||||
if err := sett.SetClaimed(); err != nil { // claimed → orphan card, NEVER auto-reset
|
||||
t.Fatal(err)
|
||||
}
|
||||
var events []string
|
||||
m.SetOffboxOrphanEvent(func(evt, _ string) { events = append(events, evt) })
|
||||
m.SetOffboxRunner(wrongPwRunner(nil))
|
||||
|
||||
if err := m.RunOffboxBackup(context.Background()); err != nil {
|
||||
t.Fatalf("run should skip cleanly on an orphaned repo, got %v", err)
|
||||
}
|
||||
if !m.OffboxOrphaned() {
|
||||
t.Fatal("repo was not classified/persisted as ORPHANED")
|
||||
}
|
||||
if len(events) != 1 || events[0] != "offbox_repo_orphaned" {
|
||||
t.Fatalf("expected exactly one offbox_repo_orphaned event, got %v", events)
|
||||
}
|
||||
got := sett.GetOffboxTarget()
|
||||
if got.RepoState != "orphaned" || got.OrphanedAt == "" {
|
||||
t.Fatalf("RepoState=%q OrphanedAt=%q, want orphaned + a stamp", got.RepoState, got.OrphanedAt)
|
||||
}
|
||||
// The raw restic error must NOT be surfaced as the last-error banner (the card explains instead).
|
||||
if strings.Contains(got.LastError, "wrong password") || strings.Contains(got.LastError, "exit status") {
|
||||
t.Fatalf("raw restic error leaked into LastError: %q", got.LastError)
|
||||
}
|
||||
// A second scheduled run SKIPS (no nightly spam) — no new event.
|
||||
if err := m.RunOffboxBackup(context.Background()); err != nil {
|
||||
t.Fatalf("second run: %v", err)
|
||||
}
|
||||
if len(events) != 1 {
|
||||
t.Fatalf("nightly re-fire — events=%v, want the single transition event only", events)
|
||||
}
|
||||
}
|
||||
|
||||
// Scenario B: an UNCLAIMED box auto-resets on detection — move-aside (never delete) + re-init; both
|
||||
// events fire and the box ends un-orphaned (next run green).
|
||||
func TestOffbox_OrphanDetection_UnclaimedAutoReset(t *testing.T) {
|
||||
m, sett := newOffboxManager(t) // unclaimed by default
|
||||
var events []string
|
||||
m.SetOffboxOrphanEvent(func(evt, _ string) { events = append(events, evt) })
|
||||
var sshCmds []string
|
||||
m.SetOffboxSSH(func(_ context.Context, _, _ string, _ int, _, _, remoteCmd string) ([]byte, error) {
|
||||
sshCmds = append(sshCmds, remoteCmd)
|
||||
if strings.HasPrefix(remoteCmd, "test -e") {
|
||||
return nil, fmt.Errorf("exit status 1") // absent → free name
|
||||
}
|
||||
return nil, nil // mv OK
|
||||
})
|
||||
m.SetOffboxRunner(wrongPwRunner(nil))
|
||||
|
||||
if err := m.RunOffboxBackup(context.Background()); err != nil {
|
||||
t.Fatalf("unclaimed run should auto-reset + succeed, got %v", err)
|
||||
}
|
||||
if m.OffboxOrphaned() {
|
||||
t.Fatal("unclaimed box stayed orphaned — auto-reset did not clear the state")
|
||||
}
|
||||
got := sett.GetOffboxTarget()
|
||||
if got.OrphanedRenamedTo == "" || !strings.Contains(got.OrphanedRenamedTo, ".orphaned-") {
|
||||
t.Fatalf("move-aside path not recorded: %q", got.OrphanedRenamedTo)
|
||||
}
|
||||
var mvSeen bool
|
||||
for _, c := range sshCmds {
|
||||
if strings.HasPrefix(c, "mv ") {
|
||||
mvSeen = true
|
||||
}
|
||||
}
|
||||
if !mvSeen {
|
||||
t.Fatalf("no move-aside mv issued: %v", sshCmds)
|
||||
}
|
||||
// Both transition events fired (orphaned → reset). No delete anywhere.
|
||||
if len(events) != 2 || events[0] != "offbox_repo_orphaned" || events[1] != "offbox_repo_reset" {
|
||||
t.Fatalf("events = %v, want [orphaned reset]", events)
|
||||
}
|
||||
}
|
||||
|
||||
// Scenario C: the claimed confirmed reset (ResetOrphanedRepo) refuses unless orphaned, then move-aside +
|
||||
// re-init + clear state.
|
||||
func TestOffbox_ConfirmedReset(t *testing.T) {
|
||||
m, sett := newOffboxManager(t)
|
||||
if err := sett.SetClaimed(); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
// refuse when not orphaned
|
||||
if err := m.ResetOrphanedRepo(context.Background()); err == nil {
|
||||
t.Fatal("reset must refuse when the repo is not orphaned")
|
||||
}
|
||||
// mark orphaned, then confirm reset
|
||||
m.SetOffboxRunner(wrongPwRunner(nil))
|
||||
_ = m.RunOffboxBackup(context.Background())
|
||||
if !m.OffboxOrphaned() {
|
||||
t.Fatal("precondition: not orphaned")
|
||||
}
|
||||
var mv bool
|
||||
m.SetOffboxSSH(func(_ context.Context, _, _ string, _ int, _, _, cmd string) ([]byte, error) {
|
||||
if strings.HasPrefix(cmd, "test -e") {
|
||||
return nil, fmt.Errorf("exit 1")
|
||||
}
|
||||
if strings.HasPrefix(cmd, "mv ") {
|
||||
mv = true
|
||||
}
|
||||
return nil, nil
|
||||
})
|
||||
if err := m.ResetOrphanedRepo(context.Background()); err != nil {
|
||||
t.Fatalf("confirmed reset: %v", err)
|
||||
}
|
||||
if !mv {
|
||||
t.Fatal("confirmed reset did not move the old repo aside")
|
||||
}
|
||||
if m.OffboxOrphaned() {
|
||||
t.Fatal("state not cleared after confirmed reset")
|
||||
}
|
||||
}
|
||||
@@ -168,6 +168,18 @@ type OffboxTarget struct {
|
||||
// hub-verified pending→escrowed flip (report.EscrowAutoConfirmer). Zeroed by that flip (and the
|
||||
// deprecated manual confirm). Persisted, so it survives a controller restart mid-wait. Not a secret.
|
||||
CeremonyCompletedAt string `json:"ceremony_completed_at,omitempty"`
|
||||
|
||||
// RepoState (v0.142.0, offsite continuity) classifies the offsite REPO — "" normal | "orphaned".
|
||||
// ORPHANED = the remote repo exists but was keyed under a passphrase this controller no longer has
|
||||
// (the reinstall/recreated-volume shape: `restic cat config` → "wrong password or no key found").
|
||||
// While orphaned, scheduled runs SKIP (one event, not nightly) and the remote page shows the orphan
|
||||
// card instead of the raw restic error; a reset (move-aside + init) clears it. Not a secret.
|
||||
RepoState string `json:"repo_state,omitempty"`
|
||||
// OrphanedAt is the RFC3339 stamp of the orphan detection (drives the card copy).
|
||||
OrphanedAt string `json:"orphaned_at,omitempty"`
|
||||
// OrphanedRenamedTo records the move-aside path of the last reset (e.g. <repo>.orphaned-20260717),
|
||||
// so the card/log can name where the old (recovery-code-recoverable) history was set aside.
|
||||
OrphanedRenamedTo string `json:"orphaned_renamed_to,omitempty"`
|
||||
}
|
||||
|
||||
// CrossDriveBackup configures per-app backup to a secondary drive.
|
||||
|
||||
@@ -755,6 +755,8 @@ func (s *Server) backupsRemoteHandler(w http.ResponseWriter, r *http.Request) {
|
||||
awaiting, timedOut := offboxCeremonyWaitState(s.settings.GetOffboxTarget())
|
||||
data["OffboxCeremonyAwaiting"] = awaiting
|
||||
data["OffboxCeremonyTimedOut"] = timedOut
|
||||
// v0.142.0 offsite-repo continuity: the orphan card + the auto-refresh (Part C) trigger.
|
||||
data["OffboxOrphaned"] = s.backupMgr != nil && s.backupMgr.OffboxOrphaned()
|
||||
s.executeTemplate(w, r, "backups_remote", data)
|
||||
}
|
||||
|
||||
|
||||
@@ -3,6 +3,7 @@ package web
|
||||
import (
|
||||
"bytes"
|
||||
"context"
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"io"
|
||||
"log"
|
||||
@@ -70,6 +71,54 @@ func TestOffboxWeb_RunGatedUntilConfirm(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
// Part C — the status endpoint (poll source) reports the current run status/snapshots as JSON.
|
||||
func TestOffboxStatusHandler(t *testing.T) {
|
||||
s, sett, _ := newOffboxWebServer(t)
|
||||
if err := sett.SetOffboxTarget(&settings.OffboxTarget{
|
||||
Enabled: true, Host: "nas.local", User: "felhom", RepoPath: "/srv/repo",
|
||||
LastStatus: "running", SnapshotCount: 7,
|
||||
}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
w := httptest.NewRecorder()
|
||||
s.offboxStatusHandler(w, httptest.NewRequest("GET", "/backup/offbox/status", nil))
|
||||
var d map[string]any
|
||||
if err := json.Unmarshal(w.Body.Bytes(), &d); err != nil {
|
||||
t.Fatalf("decode: %v (%s)", err, w.Body.String())
|
||||
}
|
||||
if d["status"] != "running" {
|
||||
t.Fatalf("status = %v, want running", d["status"])
|
||||
}
|
||||
if d["snapshots"].(float64) != 7 {
|
||||
t.Fatalf("snapshots = %v, want 7", d["snapshots"])
|
||||
}
|
||||
if d["orphaned"] != false {
|
||||
t.Fatalf("orphaned = %v, want false", d["orphaned"])
|
||||
}
|
||||
}
|
||||
|
||||
// Part A edge — an ORPHANED repo routes "Távoli mentés most" to the card, never attempting the write.
|
||||
func TestOffboxRun_RefusedWhenOrphaned(t *testing.T) {
|
||||
s, sett, m := newOffboxWebServer(t)
|
||||
if err := m.WriteOffboxSecrets("KEYMATERIAL", "nas.local ssh-ed25519 HOSTKEY"); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := sett.SetOffboxTarget(&settings.OffboxTarget{
|
||||
Enabled: true, Host: "nas.local", Port: 22, User: "felhom", RepoPath: "/srv/repo", Schedule: "daily",
|
||||
EscrowState: "escrowed", RepoState: "orphaned",
|
||||
}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
w := httptest.NewRecorder()
|
||||
s.offboxRunHandler(w, httptest.NewRequest("POST", "/backup/offbox/run", nil))
|
||||
if w.Code != 302 {
|
||||
t.Fatalf("orphaned run must redirect, got %d", w.Code)
|
||||
}
|
||||
if loc := w.Header().Get("Location"); !strings.Contains(loc, "el%C3%A1rvult") {
|
||||
t.Fatalf("orphaned run must redirect to the orphan-card flash, got %q", loc)
|
||||
}
|
||||
}
|
||||
|
||||
// Scenario E — the confirm flip wipes the agent-staged secret; a wipe failure is logged loudly but does
|
||||
// NOT fail the confirm (the state flip is the primary effect).
|
||||
func TestOffboxWeb_ConfirmWipesStagedSecret(t *testing.T) {
|
||||
|
||||
@@ -2,6 +2,7 @@ package web
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/json"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"strconv"
|
||||
@@ -209,6 +210,12 @@ func (s *Server) offboxRunHandler(w http.ResponseWriter, r *http.Request) {
|
||||
offboxRedirect(w, r, "A távoli mentés a kulcs letétbe helyezésére vár.", true)
|
||||
return
|
||||
}
|
||||
// Offsite-repo continuity (v0.142.0): an ORPHANED repo can't be written — route the customer to the
|
||||
// orphan card's explanation/reset instead of attempting a doomed write.
|
||||
if s.backupMgr.OffboxOrphaned() {
|
||||
offboxRedirect(w, r, "A távoli tároló elárvult — előbb indíts új távoli mentést a kártyán látható módon.", true)
|
||||
return
|
||||
}
|
||||
go func() {
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 3*time.Hour)
|
||||
defer cancel()
|
||||
@@ -219,6 +226,50 @@ func (s *Server) offboxRunHandler(w http.ResponseWriter, r *http.Request) {
|
||||
offboxRedirect(w, r, "A távoli mentés elindult (a futás után az állapot frissül).", false)
|
||||
}
|
||||
|
||||
// offboxResetHandler is the CLAIMED confirmed orphaned-repo reset (Scenario C): move the old (recovery-
|
||||
// code-recoverable) history aside — never delete — and init a fresh repo. Refuses unless orphaned AND
|
||||
// explicitly confirmed (confirm=1, set by the reveal-then-confirm block on the orphan card).
|
||||
func (s *Server) offboxResetHandler(w http.ResponseWriter, r *http.Request) {
|
||||
if s.backupMgr == nil || !s.backupMgr.OffboxConfigured() {
|
||||
offboxRedirect(w, r, "A távoli mentési cél nincs beállítva.", true)
|
||||
return
|
||||
}
|
||||
if !s.backupMgr.OffboxOrphaned() {
|
||||
offboxRedirect(w, r, "Az offsite tároló nincs elárvult állapotban.", true)
|
||||
return
|
||||
}
|
||||
if r.FormValue("confirm") != "1" {
|
||||
offboxRedirect(w, r, "A visszaállításhoz megerősítés szükséges.", true)
|
||||
return
|
||||
}
|
||||
go func() {
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 5*time.Minute)
|
||||
defer cancel()
|
||||
if err := s.backupMgr.ResetOrphanedRepo(ctx); err != nil {
|
||||
s.logger.Printf("[WARN] [web] offbox orphaned-repo reset failed: %v", err)
|
||||
}
|
||||
}()
|
||||
offboxRedirect(w, r, "Új távoli mentés indítása folyamatban — a régi előzmény félretéve (nem törölve).", false)
|
||||
}
|
||||
|
||||
// offboxStatusHandler (Part C) is the poll source for the remote-backup run status — the page polls it
|
||||
// after "Távoli mentés most" and flips to the terminal state without a manual reload. Session-auth'd.
|
||||
func (s *Server) offboxStatusHandler(w http.ResponseWriter, r *http.Request) {
|
||||
t := s.settings.GetOffboxTarget()
|
||||
resp := map[string]any{"status": "", "snapshots": 0, "orphaned": false}
|
||||
if t != nil {
|
||||
resp["status"] = t.LastStatus
|
||||
resp["snapshots"] = t.SnapshotCount
|
||||
resp["last_run"] = t.LastRun
|
||||
resp["last_duration"] = t.LastDuration
|
||||
resp["repo_size_human"] = t.RepoSizeHuman
|
||||
resp["last_error"] = t.LastError
|
||||
resp["orphaned"] = t.RepoState == "orphaned"
|
||||
}
|
||||
w.Header().Set("Content-Type", "application/json")
|
||||
_ = json.NewEncoder(w).Encode(resp)
|
||||
}
|
||||
|
||||
// offboxRestoreHandler restores an app's off-box data to an on-data-drive scratch dir (§7, F-A1;
|
||||
// non-destructive — does NOT overwrite live data). mode=unit (default) restores the recovery unit
|
||||
// only; mode=full is size-gated and two-step (first POST computes the size + headroom and redirects
|
||||
|
||||
@@ -392,6 +392,10 @@ func (s *Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
|
||||
s.offboxToggleHandler(w, r)
|
||||
case path == "/backup/offbox/run" && r.Method == http.MethodPost:
|
||||
s.offboxRunHandler(w, r)
|
||||
case path == "/backup/offbox/reset" && r.Method == http.MethodPost:
|
||||
s.offboxResetHandler(w, r)
|
||||
case path == "/backup/offbox/status" && r.Method == http.MethodGet:
|
||||
s.offboxStatusHandler(w, r)
|
||||
case path == "/backup/offbox/restore" && r.Method == http.MethodPost:
|
||||
s.offboxRestoreHandler(w, r)
|
||||
case path == "/backup/offbox/place" && r.Method == http.MethodPost:
|
||||
|
||||
@@ -28,7 +28,7 @@
|
||||
{{if .Offbox}}
|
||||
<div class="stats-grid backup-page-cards">
|
||||
<div class="stat-card {{if eq .Offbox.LastStatus "error"}}stat-warn{{end}}">
|
||||
<div class="stat-value" style="font-size:1.15rem">{{if eq .Offbox.LastStatus "ok"}}✓ Rendben{{else if eq .Offbox.LastStatus "error"}}✗ Hiba{{else if eq .Offbox.LastStatus "running"}}Fut…{{else}}–{{end}}</div>
|
||||
<div class="stat-value" id="offbox-status-value" style="font-size:1.15rem">{{if eq .Offbox.LastStatus "ok"}}✓ Rendben{{else if eq .Offbox.LastStatus "error"}}✗ Hiba{{else if eq .Offbox.LastStatus "running"}}Fut…{{else}}–{{end}}</div>
|
||||
<div class="stat-label">Utolsó távoli mentés{{if .Offbox.LastRun}}<br><span class="relative-time">{{timeAgoStr .Offbox.LastRun}}</span>{{end}}</div>
|
||||
</div>
|
||||
<div class="stat-card">
|
||||
@@ -53,6 +53,23 @@
|
||||
{{/* Part E: display pick — a stale zero-toggle warning is replaced once the selection
|
||||
changed (neutral color: the replacement is reassurance, not a deviation). */}}
|
||||
{{if .OffboxWarningDisplay}}<p class="form-hint"{{if eq .OffboxWarningDisplay .Offbox.LastWarning}} style="color:var(--warn)"{{end}}>{{.OffboxWarningDisplay}}</p>{{end}}
|
||||
{{/* v0.142.0 offsite-repo continuity: the ORPHANED card replaces the raw restic error banner. The
|
||||
remote holds backups written under a previous, no-longer-available key (reinstall shape). A
|
||||
reset moves the old history aside (never deletes) and starts a fresh repo under the current key. */}}
|
||||
{{if eq .Offbox.RepoState "orphaned"}}
|
||||
<div class="card" style="border-left:3px solid var(--crit,#e5484d);margin:.75rem 0;padding:.75rem 1rem" id="offbox-orphan-card">
|
||||
<p style="margin:0 0 .35rem;font-weight:600">A távoli tároló másik kulccsal készült mentéseket tartalmaz</p>
|
||||
<p class="form-hint" style="margin:0 0 .5rem">A távoli tárhelyen lévő mentések egy korábbi, már nem elérhető kulccsal készültek (jellemzően újratelepítés után). Emiatt új mentés jelenleg nem írható a tárolóba. A meglévő mentések nem sérültek — a hozzájuk tartozó helyreállítási kóddal később visszaállíthatók lehetnek.</p>
|
||||
<button type="button" class="btn btn-sm btn-outline" id="orphan-reveal" onclick="var c=document.getElementById('orphan-confirm');c.style.display='block';this.style.display='none'">Új távoli mentés indítása…</button>
|
||||
<div id="orphan-confirm" style="display:none;margin-top:.6rem">
|
||||
<p class="form-hint" style="margin:0 0 .5rem">A régi előzmény <strong>félretéve marad</strong> (nem törlődik), és a hozzá tartozó helyreállítási kóddal később visszaállítható lehet. Egy üres, új tároló jön létre a mostani kulccsal, és a következő mentés ide készül.</p>
|
||||
<form method="POST" action="/backup/offbox/reset" style="display:inline">{{.CSRFField}}
|
||||
<input type="hidden" name="confirm" value="1">
|
||||
<button type="submit" class="btn btn-sm btn-primary">Megerősítés — új távoli mentés indítása</button>
|
||||
</form>
|
||||
</div>
|
||||
</div>
|
||||
{{end}}
|
||||
{{/* Escrow ceremony card (v0.127.0): the customer-driveable wizard replaced the manual-confirm
|
||||
button (that deprecated endpoint stays for legacy blobs; its button is gone). States:
|
||||
awaiting (v0.138.0: ceremony done, hub confirm pending) → info; awaiting timed out → warn +
|
||||
@@ -154,5 +171,25 @@
|
||||
</div>
|
||||
{{end}}
|
||||
|
||||
<script>
|
||||
/* Part C (v0.142.0): while a remote backup is in flight the status card shows "Fut…" — poll the run
|
||||
status and reload once it reaches a terminal state, so the customer sees Rendben/Hiba + fresh
|
||||
numbers WITHOUT a manual reload. Polls only when a run is running; stops (page reload) at terminal. */
|
||||
(function(){
|
||||
var v=document.getElementById('offbox-status-value');
|
||||
if(!v || v.textContent.indexOf('Fut')<0) return; // not running → nothing to poll
|
||||
var timer=setInterval(function(){
|
||||
fetch('/backup/offbox/status',{headers:{'Accept':'application/json'}})
|
||||
.then(function(r){return r.json();})
|
||||
.then(function(d){
|
||||
if(d && d.status==='running') return; // still running → keep polling
|
||||
clearInterval(timer);
|
||||
location.reload(); // terminal → re-render (fresh numbers, warnings, or the orphan card)
|
||||
})
|
||||
.catch(function(){ /* transient — keep polling */ });
|
||||
}, 3000);
|
||||
})();
|
||||
</script>
|
||||
|
||||
{{template "layout_end" .}}
|
||||
{{end}}
|
||||
|
||||
Reference in New Issue
Block a user