Keep the local index when remote nuke cannot list or find the destination #384

Merged
clawbot merged 1 commits from issue-292-nuke-keeps-local-index into next 2026-10-14 06:47:40 +02:00
8 changed files with 142 additions and 37 deletions
+6 -5
View File
@@ -385,11 +385,12 @@ bucket/
A destination without `vaultik-destination.txt` is treated as not present
while the local index holds snapshots stored there (`checkDestinationMarker`
in `internal/vaultik/destination_marker.go`): `snapshot create` refuses to
write to it, `prune` and `snapshot purge` fail, and `snapshot list` and
`snapshot remove` behave as for a destination that cannot be listed. This
keeps those commands from taking an unplugged volume's empty mount point for
a destination whose snapshots are all gone. The `file://` backend never
creates its destination directory, so a write to a missing one fails.
write to it, `prune`, `snapshot purge` and `remote nuke` fail, and
`snapshot list` and `snapshot remove` behave as for a destination that cannot
be listed. This keeps those commands from taking an unplugged volume's empty
mount point for a destination whose snapshots are all gone. The `file://`
backend never creates its destination directory, so a write to a missing one
fails.
The `{remote-key}` directory name is a one-way double SHA-256 hash of the human
snapshot ID, so the human ID (hostname, snapshot name, timestamp) is never
+13 -7
View File
@@ -454,9 +454,14 @@ since `prune` never deletes it.
the backup destination store, whatever its name, and then
`vaultik-destination.txt`, leaving the bucket prefix empty. Destructive
and irreversible. This is the single supported way to wipe the entire
destination store. If a listing afterwards still finds anything under
`metadata/` or `blobs/`, the command names each object left and exits
non-zero; running it again deletes what is left.
destination store. A snapshot is removed from the local index only once
its metadata is deleted from the destination store. If the destination
store cannot be listed, or the local index holds snapshots and the
destination has no `vaultik-destination.txt` (an unplugged volume's
empty mount point), the command fails before it removes anything. If a
listing afterwards still finds anything under `metadata/` or `blobs/`,
the command names each object left and exits non-zero; running it again
deletes what is left.
* `--force`: Required to confirm destruction.
**`database delete`**: Delete the local SQLite state database file
@@ -529,10 +534,11 @@ For an implementation-level view of the internals — the data model, the
* `vaultik-destination.txt` holds one fixed line of text, with no host
name, path or time. A destination without it is treated as not
present while the local index holds snapshots stored there:
`snapshot create` refuses to write to it, `prune` and `snapshot purge`
fail, and `snapshot list` and `snapshot remove` behave as for a
destination that cannot be listed. This keeps an unplugged volume's empty mount point
from being taken for a destination whose snapshots are all gone. Mount
`snapshot create` refuses to write to it, `prune`, `snapshot purge`
and `remote nuke` fail, and `snapshot list` and `snapshot remove`
behave as for a destination that cannot be listed. This keeps an
unplugged volume's empty mount point from being taken for a
destination whose snapshots are all gone. Mount
the volume if it is removable; if the destination was emptied on
purpose, run `vaultik database delete` and back up again.
* Blobs are two-level directory sharded using the first 4 hex chars of the blob hash
+1 -1
View File
@@ -27,7 +27,7 @@ The metadata subdirectory is named with the **remote key**, a one-way hash of th
## Marker File (`vaultik-destination.txt`)
Every backup writes this file at the root before it uploads anything. It holds one fixed line of plain text and names no host, path or time. A destination without it is treated as not present while the local index holds snapshots stored there: `snapshot create` refuses to write to it, and `prune`, `snapshot purge`, `snapshot remove` and `snapshot list` behave as for a destination that cannot be listed. An unplugged volume leaves an empty directory at its mount point, and without the marker that directory would read as a destination whose snapshots are all gone. `remote nuke` deletes the marker last.
Every backup writes this file at the root before it uploads anything. It holds one fixed line of plain text and names no host, path or time. A destination without it is treated as not present while the local index holds snapshots stored there: `snapshot create` refuses to write to it, and `prune`, `snapshot purge`, `snapshot remove`, `snapshot list` and `remote nuke` behave as for a destination that cannot be listed. An unplugged volume leaves an empty directory at its mount point, and without the marker that directory would read as a destination whose snapshots are all gone. `remote nuke` deletes the marker last.
## Blobs Directory (`blobs/`)
@@ -152,6 +152,25 @@ func TestPurgeRefusesDestinationWithoutMarker(t *testing.T) {
assertUntouched(ctx, t, repos, storeDir)
}
// TestNukeRemoteRefusesDestinationWithoutMarker checks that remote nuke
// fails, deletes no local snapshot record, and neither reports the
// destination empty nor advises running nuke again when the destination
// has no marker.
func TestNukeRemoteRefusesDestinationWithoutMarker(t *testing.T) {
log.Initialize(log.Config{})
t.Parallel()
ctx := context.Background()
v, repos, out, storeDir := backUpThenUnplugMountPoint(ctx, t)
err := v.NukeRemote(true)
require.ErrorContains(t, err, markerFile)
assert.NotContains(t, out.String(), "now empty")
assert.NotContains(t, out.String(), "vaultik remote nuke --force")
assertUntouched(ctx, t, repos, storeDir)
}
// TestRemoveSnapshotWarnsWhenDestinationHasNoMarker checks that snapshot
// remove warns that the metadata could not be removed from a destination
// without a marker, instead of reporting that it was.
@@ -188,3 +188,41 @@ func TestPurgeSaysListingFailedOnceWhenDestinationMissing(t *testing.T) {
assert.Equal(t, 1, strings.Count(err.Error(), "listing remote snapshots"),
err.Error())
}
// TestNukeRemoteKeepsLocalIndexWhenDestinationMissing checks that remote
// nuke fails on a destination it cannot list, leaves the local index
// unchanged, and neither reports the destination empty nor advises
// running nuke again.
func TestNukeRemoteKeepsLocalIndexWhenDestinationMissing(t *testing.T) {
log.Initialize(log.Config{})
t.Parallel()
ctx := context.Background()
v, repos, out := backUpThenUnplug(ctx, t)
filesBefore, err := repos.Files.ListAll(ctx)
require.NoError(t, err)
require.NotEmpty(t, filesBefore)
blobsBefore, err := repos.Blobs.GetAll(ctx)
require.NoError(t, err)
require.NotEmpty(t, blobsBefore)
err = v.NukeRemote(true)
require.ErrorIs(t, err, fs.ErrNotExist)
require.ErrorContains(t, err, "listing remote snapshots")
assert.NotContains(t, out.String(), "now empty")
assert.NotContains(t, out.String(), "vaultik remote nuke --force")
snapshots, err := repos.Snapshots.ListRecent(ctx, listRecentTestLimit)
require.NoError(t, err)
assert.Len(t, snapshots, 1, "nuke must delete no local snapshot record")
files, err := repos.Files.ListAll(ctx)
require.NoError(t, err)
assert.Len(t, files, len(filesBefore), "nuke must delete no local file record")
blobs, err := repos.Blobs.GetAll(ctx)
require.NoError(t, err)
assert.Len(t, blobs, len(blobsBefore), "nuke must delete no local blob record")
}
+27 -2
View File
@@ -104,6 +104,32 @@ func TestNukeRemoteFailsWhenADeleteFails(t *testing.T) {
}
}
// TestNukeRemoteKeepsLocalSnapshotWhoseMetadataIsNotDeleted checks that
// remote nuke keeps a snapshot in the local index while its metadata
// cannot be deleted from the destination store: dropped from the local
// index, it would be left there as a remote-only snapshot without its
// name.
func TestNukeRemoteKeepsLocalSnapshotWhoseMetadataIsNotDeleted(t *testing.T) {
log.Initialize(log.Config{})
t.Parallel()
ctx := context.Background()
storeDir := filepath.Join(t.TempDir(), "store")
v, repos, _ := backUpToFileDestination(ctx, t, storeDir)
store := faultstore.New(v.Storage)
store.OnDelete = func(key string) bool {
return strings.HasPrefix(key, "metadata/")
}
v.Storage = store
require.Error(t, v.NukeRemote(true))
snapshots, err := repos.Snapshots.ListRecent(ctx, listRecentTestLimit)
require.NoError(t, err)
assert.Len(t, snapshots, 1)
}
// TestNukeRemoteNamesWhatIsLeft checks that remote nuke, when an object
// it does not recognise cannot be deleted, does not report the store
// empty but names that object and fails.
@@ -154,8 +180,7 @@ func TestNukeRemoteRefusesAnotherDestinationWithoutRetryAdvice(t *testing.T) {
// TestNukeRemoteFailsWhenALocalSnapshotCannotBeRemoved checks that a
// snapshot remote nuke cannot remove from the local index fails the
// run: nuke then leaves that snapshot's metadata on the destination
// store.
// run, and that nuke then does not report the destination store empty.
func TestNukeRemoteFailsWhenALocalSnapshotCannotBeRemoved(t *testing.T) {
log.Initialize(log.Config{})
t.Parallel()
+8 -1
View File
@@ -55,12 +55,19 @@ func (v *Vaultik) NukeRemote(force bool) error {
}
// Checked before anything is deleted: running nuke again would fail
// the same way, so the advice below must not follow this error.
// the same way, so the advice below must not follow these errors.
// The listing also fails on a destination without the marker while
// the local index holds snapshots.
err := v.EnsureStorageBinding()
if err != nil {
return err
}
_, err = v.listAllRemoteSnapshotKeys()
if err != nil {
return err
}
err = v.emptyDestinationStore()
if err != nil {
v.UI.Warningf("Run '%s' again to delete what is left on the "+
+30 -21
View File
@@ -1516,7 +1516,9 @@ func (v *Vaultik) removeSnapshotRemote(
// database from the local index, and (with --remote) every snapshot
// metadata directory in remote storage. Both sides are processed so a
// "remove --all" leaves nothing behind, even when the local DB and
// remote storage have diverged.
// remote storage have diverged. Unless LocalOnly is set, it fails
// before it removes anything when the destination store cannot be
// listed or is not present by checkDestinationMarker.
func (v *Vaultik) RemoveAllSnapshots(opts *RemoveOptions) (*RemoveResult, error) {
err := v.EnsureStorageBinding()
if err != nil {
@@ -1529,11 +1531,15 @@ func (v *Vaultik) RemoveAllSnapshots(opts *RemoveOptions) (*RemoveResult, error)
}
// remoteKeys is the set of metadata/<key>/ subdirectories on the
// destination store; failures are downgraded to a warning so a
// permission-denied or unreachable remote can't block a local-only
// remove.
// destination store. Only for a local-only remove is a failure
// downgraded to a warning, so a permission-denied or unreachable
// remote can't block it.
remoteKeys, remoteErr := v.listAllRemoteSnapshotKeys()
if remoteErr != nil {
if !opts.LocalOnly {
return nil, remoteErr
}
log.Warn("Could not list remote snapshots", "error", remoteErr)
v.UI.Warningf("Could not list remote snapshots: %v.", remoteErr)
}
@@ -1595,8 +1601,7 @@ func (v *Vaultik) localSnapshotIDs() ([]string, error) {
// listAllRemoteSnapshotKeys collects the hashed remote keys
// (subdirectories under metadata/) currently present in the
// destination store. Returns (nil, err) when the store cannot be
// listed, or is not present by checkDestinationMarker; callers must
// treat that as "no remote info available," not fatal.
// listed, or is not present by checkDestinationMarker.
func (v *Vaultik) listAllRemoteSnapshotKeys() ([]string, error) {
log.Info("Listing all remote snapshots")
@@ -1780,10 +1785,12 @@ func (v *Vaultik) deleteIncompleteSnapshot(snapshotIDStr string, id any) bool {
return true
}
// removeAllLocalSnapshots removes each known local snapshot (and, unless
// --local-only, its remote metadata), returning the number of snapshots
// it could not remove. A snapshot left in the local index keeps its
// remote metadata too, since it is not an orphan remote key.
// removeAllLocalSnapshots removes each known local snapshot's remote
// metadata, unless --local-only, and then its local record, returning
// the number of snapshots it could not remove. The local record goes
// only once the metadata is deleted: dropped first, it would leave that
// metadata on the destination store as a remote-only snapshot whose name
// is lost.
func (v *Vaultik) removeAllLocalSnapshots(
localSnaps []string, opts *RemoveOptions, result *RemoveResult,
) int {
@@ -1792,6 +1799,19 @@ func (v *Vaultik) removeAllLocalSnapshots(
for _, snapshotID := range localSnaps {
log.Info("Removing snapshot", "snapshot_id", snapshotID)
if !opts.LocalOnly {
_, err := v.deleteRemoteSnapshotByKey(
snapshot.RemoteSnapshotKey(snapshotID))
if err != nil {
log.Warn("Failed to remove snapshot metadata from remote",
"snapshot_id", snapshotID, "error", err)
failures++
continue
}
}
err := v.deleteSnapshotFromLocalDB(snapshotID)
if err != nil {
log.Error("Failed to remove from local database",
@@ -1802,17 +1822,6 @@ func (v *Vaultik) removeAllLocalSnapshots(
continue
}
if !opts.LocalOnly {
_, err = v.deleteRemoteSnapshotByKey(
snapshot.RemoteSnapshotKey(snapshotID))
if err != nil {
log.Warn("Failed to remove snapshot metadata from remote",
"snapshot_id", snapshotID, "error", err)
failures++
}
}
result.SnapshotsRemoved = append(result.SnapshotsRemoved, snapshotID)
}