A backup uploads its blobs before the manifest that lists them, and the
process lock covers one user on one machine, so a prune from another host
or user sharing the destination deleted a running backup's blobs.
Prune now deletes an unreferenced blob only once its listed time is older
than the new config key prune_grace_period (default 24h; a negative value
is refused). A newer blob, or one listed with no time, is kept and counted
(blobs_too_new). A backup uploads a blob already on the destination again
instead of skipping it, so it gets a fresh time. The rclone backend takes
each object's time from the listing (use_server_modtime).
Cost: after database delete, the next backup uploads all its data again.
Side effect: a rebuilt blob replaces the stored copy; if compression_level
or the recipients changed since, older snapshots listing it fail verify.
Model: opus-5-5
`remote nuke` took a destination store it could not list, or one
without `vaultik-destination.txt` while the local index held
snapshots, as a warning: it then cleared the local snapshot records,
and for an unplugged volume's empty mount point reported the store
empty and exited zero. Nuke now fails on either before it removes
anything, without its advice to run it again, which would fail the
same way. The bulk remove it runs fails on either too; only a
local-only remove still goes on with a warning. Nuke also deletes each
snapshot's remote metadata before its local record, so a failed delete
keeps the snapshot in the local index rather than leaving it on the
store as a remote-only snapshot.
Model: opus-5-5
Every backup writes vaultik-destination.txt at the destination root.
While the local index holds snapshots stored there, a destination
without it is not present: snapshot create refuses before writing,
prune and snapshot purge fail, and snapshot list and snapshot remove
act as for a destination that cannot be listed. snapshot create and
snapshot remove check it first. remote nuke deletes the marker last.
The file:// backend never creates its destination directory, which
protects a directory on a removable volume, not a mount point.
Tests back up into directories that exist; fixtures for a written
destination carry the marker.
Judgement call: the marker is rewritten on every backup.
Deviation: for the complexity limit, two blocks moved, unchanged, into
snapshotNamesToCreate and confirmRemoveSnapshot.
Model: opus-5-5
finalizeSnapshotMetadata marked the snapshot complete and then exported its metadata. A crash after completion but before/during the export left the local index showing the snapshot complete while the destination had no manifest or database, and PruneDatabase (which drops only NULL completed_at rows) kept it: a silently unrestorable snapshot.
Reorder so completion is recorded last. CompleteSnapshot is split into PopulateSnapshotBlobs (before the export) and MarkSnapshotComplete (after it). An interrupted export now leaves the snapshot incomplete, so the next run PruneDatabase drops it and re-backs-up the data; the reverse tiny window leaves a restorable snapshot the index reports honestly as remote-only. Update REPOSTRUCTURE.md guarantee 4 and the ARCHITECTURE.md flow. Add a fault-injection test driving the full create path.
Model: opus-4-8
Docs and comments only; no behaviour change. Corrects ten overclaims the security review found: snapshot names are hashed but the hash uses no secret, so a guessed hostname and name can be confirmed; a blob is named by hex(SHA256(SHA256(uncompressed contents))), stated once in docs/REPOSTRUCTURE.md and referenced elsewhere; double hashing does not hide known content (blob packing does); age uses ChaCha20-Poly1305, not XChaCha20; encryption is required, not optional; a snapshot is marked complete before its metadata is uploaded; the export comment now matches its only caller; deep verify detects corruption, not authorship; adding a recipient does not reach existing data; restore examples target a user-owned directory.
Adds an Accepted Risks subsection under Security Considerations with the seven documented risks, cross-referenced from the README.
Model: opus-4-8
Prune learned which blobs are in use by reading every snapshot's manifest, but merely logged and skipped one it could not download or decode. Blobs referenced only by that snapshot then looked unreferenced and were deleted, with a zero exit -- and snapshot create --prune runs this unattended. collectReferencedBlobs now errors, naming the remote key, so prune deletes nothing and exits non-zero.
Manifest generation likewise skipped a blob whose lookup failed or was missing, yielding a manifest short of what the snapshot needs; it now fails. Deep verify only warned when the manifest omitted a database blob; it now fails on any divergence. Docs corrected.
Model: opus-4-8
Three documents showed the remote layout with a plaintext snapshot ID as the metadata directory name, and docs/REPOSTRUCTURE.md blamed those IDs for the observable backup time. The store actually names each metadata directory with a one-way hash of the ID, so hostname and snapshot name are not visible; the backup time is, through the plaintext timestamp in the manifest, which is accepted behaviour.
README, ARCHITECTURE.md, docs/DATAMODEL.md and docs/REPOSTRUCTURE.md now show the hashed layout, the derivation is documented once, and the privacy section lists what the unencrypted manifest exposes. Two code comments that claimed the timestamp was hidden are corrected. No behaviour change.
Judgement call: docs/DATAMODEL.md was not named in the issue but had the same error.
model: claude-opus-4-8 (implementation, review); claude-fable-5-1 (merge)
- Manifests are now only compressed (not encrypted) so pruning operations can work without private keys
- Updated generateBlobManifest to use zstd compression directly
- Updated prune command to handle unencrypted manifests
- Updated snapshot list command to handle new manifest format
- Updated documentation to reflect manifest.json.zst (not .age)
- Removed unnecessary VAULTIK_PRIVATE_KEY check from prune command
sneak
clawbot