Skip to main content
Version: v5.7.0

opschain admin database backup

Included from Basic tier

Backs up an embedded (SQLite) server into ONE manifest-sealed zip: a VACUUM INTO snapshot of the store (consistent against a running server under WAL), every file of the content-addressed blob root under blobs/, and an embedded manifest.json recording the snapshot's SHA-256, its size and the blob count. This is an offline verb run on the server host — it reads the home's config.yaml the way server start does, not the API.

For the embedded store the built-in IS the right tool; for PostgreSQL it is not — this verb refuses a PostgreSQL store by name and points at the DBA lane (quiesce, backup-info, pg_dump).

Encryption keys are NOT included: back up server/keys.env (and the vault master key) separately — losing them makes encrypted data unrecoverable.

Flags​

FlagMeaning
--out <file.zip>Write the backup to this file (default opschain-sqlite-backup-<timestamp>.zip; an existing file is never overwritten).

Exit codes​

CodeMeaning
0Backup written and sealed.
1The store is PostgreSQL (the DBA-lane refusal), the store file is missing, or writing failed.

Named errors​

  • the configured store is PostgreSQL: database byte movement belongs to the DBA's own tools — quiesce the system ..., fetch the manifest and the blessed pg_dump recipe with 'opschain admin database backup-info', take the dump and the blobstore copy, then 'opschain admin platform resume' — exit 1.

Examples​

opschain admin database backup --out nightly.zip

See also​

  • opschain docs cli/admin-database-restore — restoring the zip into an empty store.
  • opschain docs cli/admin-database-backup-info — the PostgreSQL lane.
  • opschain docs cli/server-backup — the directory-shaped predecessor this verb's zip supersedes for scripted backups.