Encrypted MongoDB backups.
xbackupman captures MongoDB with native archive dumps from mongodump, encrypts them with AES-256-GCM inside the worker, and stores the ciphertext in a bucket you own. Restores use namespace remapping so you can bring a database back beside the original without a collision.
What gets captured
- A native mongodump archive — collections, documents and indexes in MongoDB's own portable format.
- The full database (or a chosen set of collections), captured as a single streamable archive.
- In-stream gzip compression plus AES-256-GCM encryption before the archive leaves the worker.
- A SHA-256 checksum stored with every snapshot for on-demand verification.
How it works
Connect your cluster
Add a MongoDB connection string for a standalone, replica set or cluster. A read-capable user is enough to run mongodump against the databases you choose.
Native archive dump
The worker runs mongodump in --archive mode, producing one portable stream of your collections, documents and indexes rather than a directory of loose BSON files.
Encrypt and store
The archive is compressed and sealed with AES-256-GCM in the worker, then uploaded to your own S3-compatible bucket. Only ciphertext ever leaves the pipeline.
Verify and prune
Every archive is SHA-256 checksummed. Schedule it as often as you like and let retention clear old snapshots without ever deleting your final copy.
Your storage, your keys
Your MongoDB archives are written to your own bucket — S3, R2, B2 or MinIO — never to a xbackupman-owned store. Because the archive is encrypted before upload, your provider holds ciphertext only. You keep complete ownership of the data and can restore or export it from your own storage whenever you want.
xbackupman never stores your backups on its own servers. It orchestrates the pipeline and records metadata — the encrypted artifacts live only in the S3-compatible bucket you connect. See how encryption works.
One-click restore
Restores replay the mongodump archive onto a target with safe namespace remapping, so a recovered database can sit alongside the original instead of overwriting it. Pick a snapshot, confirm, and the worker decrypts and applies it — or download the archive and run mongorestore yourself for full manual control.
Questions
QWhy an archive dump instead of a directory of BSON files?
A single mongodump --archive stream is simpler to encrypt, compress and upload atomically, and it restores cleanly with mongorestore. There are no loose files to lose or reassemble.
QWhat is namespace remapping on restore?
Restores can map the archive's database and collection names to new ones, so you can bring a backup back next to the live database — into, say, myapp_restore — without clobbering production.
QDoes it work with replica sets and clusters?
Yes. mongodump reads from a connection string that can point at a standalone, a replica set or a cluster. Point it at a secondary to keep backup load off the primary.
QAre indexes included?
Yes. The archive captures index definitions alongside your documents, so a restore rebuilds the collections with their indexes intact.
Back up more
One engine covers every source you run — on a single schedule, in one dashboard, encrypted into the same storage.