Your PostgreSQL data deserves a second chance.
Acrynite is an open-source CLI for PostgreSQL backup and recovery. Back up with pg_dump or pg_basebackup to Amazon S3 or local disk, schedule runs, archive WAL and recover to a chosen point in time.
$curl -fsSL https://raw.githubusercontent.com/bendev-w/acrynite-cli/main/install.sh | sh
Illustrative session. Commands from the Acrynite CLI docs, output abbreviated.
Backups are not enough. Recovery matters.
A developer can accidentally delete critical records, a deployment can introduce a problem, a server can fail, or database corruption can occur. Having a backup is only part of the equation. Being able to recover that data when it matters is what counts.
Accidental data deletion
A mistyped DELETE or a bad migration removes records you still need.
Failed deployments
A release ships a change that leaves your data in a state you cannot keep.
Database corruption
Storage or software faults damage data files that queries depend on.
Server failures
Hardware, disks or whole hosts fail, taking the running cluster with them.
Acrynite is designed to make PostgreSQL backup and recovery easier to manage.
Tools to protect your PostgreSQL data.
What the current release supports today. Anything experimental or planned is labelled as such.
Physical backups
Capture a full PostgreSQL cluster with pg_basebackup for physical restoration.
Logical backups
Create pg_dump backups for portability, migration and restore into a staging profile.
WAL archiving
Configure archive_command so PostgreSQL write-ahead log segments are preserved.
Point-in-time recovery
Recover a physical backup to a target time in an isolated Docker container, when the matching WAL is available.
Scheduled backups
Run recurring backups on an interval such as 24h. Each run gets its own ID and manifest.
Storage flexibility
Send backups to Amazon S3 or an S3-compatible endpoint, or to a local directory for one-off runs.
A straightforward approach to database protection.
Register your database
Save a PostgreSQL connection profile once and test it.
acrynite database register production …Configure storage
Register an S3 or S3-compatible destination for backups.
acrynite storage register s3-prod-backups …Create backups
Run a logical or physical backup now, or create a schedule.
acrynite backup run --type physical …Restore when necessary
Plan, then restore into Docker with WAL replay for point-in-time recovery.
acrynite restore run --target docker …Acrynite runs when you or a schedule call it. It does not monitor your database continuously, and it never restores over the source cluster.
Watch the messages and files move.
A backup, a WAL segment and a restore, traced between your database, the Acrynite CLI, S3 and local storage.
- waiting for first event…
Built with developers in mind.
Open source
Read the code, file issues and send changes. Nothing is hidden behind a service.
Command-line driven
Every operation is a CLI command you can script, review and automate.
Runs in your environment
Acrynite runs where you run it. Backups go only to storage you configure.
Different backup approaches
Pick logical dumps, physical base backups with WAL, or both.
Recovery comes first
Restore plans, isolated recovery containers and a refusal to overwrite your source.
Better database recovery starts with the community.
Acrynite is being developed in the open. Explore the source code, report bugs, test recovery scenarios, suggest improvements and contribute to the project.
Explore GitHub →PostgreSQL backup questions, answered.
What people ask before they trust a tool with their database.
What is Acrynite?
Acrynite is an open-source command-line tool for backing up and recovering PostgreSQL databases. It creates logical backups with pg_dump and physical backups with pg_basebackup, stores them in Amazon S3, S3-compatible storage or a local directory, and restores them into an isolated Docker container.
How do I back up a PostgreSQL database to Amazon S3?
Register the database with acrynite database register, register an S3 bucket with acrynite storage register --provider s3, then run acrynite backup run --database <name> --type logical --storage <storage-name>. Use --type physical for a full cluster backup with pg_basebackup.
What is the difference between a logical and a physical PostgreSQL backup?
A logical backup exports database objects and data with pg_dump. It is portable and is restored into a separate PostgreSQL database with psql. A physical backup copies the cluster's data files with pg_basebackup. Combined with a continuous WAL archive, a physical backup supports point-in-time recovery.
Does Acrynite support PostgreSQL point-in-time recovery (PITR)?
Yes, for physical backups with WAL archiving enabled. Run acrynite restore run --backup-id <id> --target docker --target-time <UTC timestamp> to replay archived WAL into an isolated Docker container. Recovery needs a compatible base backup and an unbroken WAL sequence up to the target time.
Can I schedule automatic PostgreSQL backups?
Yes. acrynite schedule create runs a backup on an interval such as 24h or 6h. The scheduler process must keep running, and every scheduled run gets its own backup ID and manifest.
Which databases and storage providers does Acrynite support?
Acrynite supports PostgreSQL. Backups can go to Amazon S3, S3-compatible object storage or a local directory. Physical backups are restored into Docker, and logical backups into a separately registered PostgreSQL database.
How do I install Acrynite?
On Linux and macOS run the install script with curl. On Windows run the PowerShell installer. You can also download a release archive from GitHub. No Go toolchain is needed unless you build from source.
Is Acrynite open source?
Yes. The source code is on GitHub at bendev-w/acrynite-cli. Acrynite runs in your own environment and does not need a hosted service.
Start protecting your PostgreSQL databases.
Explore the documentation to install Acrynite, configure your first database, and create a backup.
Read about how Acrynite handles credentials and backup data before pointing it at production.