Postgres S3 backups

A simple NodeJS application to backup your PostgreSQL database to S3 via a cron.

Configuration

• AWS_ACCESS_KEY_ID: AWS access key ID.
• AWS_SECRET_ACCESS_KEY: AWS secret access key.
• AWS_S3_BUCKET: Name of the S3 bucket to store backups.
• AWS_S3_REGION: AWS region for the S3 bucket. Default: eu-west-2.
• BACKUP_DATABASE_URL: Connection string for the database to back up.
• BACKUP_CRON_SCHEDULE: Cron schedule for backups. Default: 0 5 * * *.
• AWS_S3_ENDPOINT: Custom S3 endpoint (e.g., for MinIO, Cloudflare R2, Backblaze B2). Default: empty.
• AWS_S3_FORCE_PATH_STYLE: Use path-style URLs for S3 (needed for some endpoints like MinIO). Default: false.
• RUN_ON_STARTUP: Run a backup immediately on startup. Default: false.
• SINGLE_SHOT_MODE: Run a single backup on start and exit. Default: true.
• SUPPORT_OBJECT_LOCK: Enable S3 Object Lock support (adds MD5 hash). Default: true.
• BACKUP_OPTIONS: Extra pg_dump options. Default: empty.
• BACKUP_PROJECT_NAME: Product name for backup pathing (e.g., my-project). Required.
• BACKUP_ENV: Environment name for backup pathing (e.g., staging, uat, prod). Required.
• BACKUP_FREQUENCY: Frequency for backup pathing (e.g., daily, weekly, monthly). Required.

See pg_dump docs for valid options for BACKUP_OPTIONS.