Supported engines
Database engines and tools Backup uses to take dumps.
Backup runs the official command-line tools pg_dump and mongodump
inside the backup worker container. The output is streamed directly to
S3.
PostgreSQL
- Tool:
pg_dump. - Format: custom format (
--format=custom,-Fc) – the formatpg_restorereads, with support for parallel restore. - Flags:
--no-owner --no-aclso dumps are restorable to a database with a different role layout than the source.
For PostgreSQL version compatibility, check the Backup service's release notes or contact us via the dashboard.
MongoDB
- Tool:
mongodump. - Format: archive (
--archive) – a single-file BSON-based stream suitable formongorestore --archive.
For MongoDB version compatibility, check the Backup service's release notes or contact us via the dashboard.
Compression
Backups are gzip-compressed in the streaming pipe between the dump tool and S3. Compression ratios depend heavily on data – text-heavy schemas compress well; tables of opaque binary blobs much less.
To configure compression per policy, talk to us via the dashboard – the option exists on the backend but isn't exposed in the UI yet.
What's not supported
- Postgres physical (
pg_basebackup) - Postgres PITR via WAL archiving
- MongoDB physical snapshots
- Per-collection or partial dumps
- Custom
pg_dumpflags (schema filters, etc.)
Reach out via app.norcube.com if any of these block your use case.
Database user requirements
For Postgres, the role needs read access to everything you want backed
up – at minimum pg_read_all_data. For MongoDB, the built-in read
role on the target database is the minimum.
See Manage credentials for SQL snippets.