Overview

The template uses node-cron or Railway cron, written in TypeScript to dump your PostgreSQL data to a file and then upload the file to S3.

Highlights

• Configurable backup schedule: By default, the cron runs at 5 AM every day but is configurable via the BACKUP_CRON_SCHEDULE environment variable.

• Support for custom buckets: The script also supports using a AWS_S3_ENDPOINT environment variable to use any S3 compliant storage bucket (eg: Wasabi).

Configuration

• AWS_ACCESS_KEY_ID - AWS access key ID.

• AWS_SECRET_ACCESS_KEY - AWS secret access key, sometimes also called an application key.

• AWS_S3_BUCKET - The name of the bucket that the access key ID and secret access key are authorized to access.

• AWS_S3_REGION - The name of the region your bucket is located in, set to auto if unknown.

• BACKUP_DATABASE_URL - The connection string of the database to backup.

• BACKUP_CRON_SCHEDULE - The cron schedule to run the backup on. Example: 0 5 * * *

• AWS_S3_ENDPOINT - The S3 custom endpoint you want to use. Applicable for 3-rd party S3 services such as Cloudflare R2 or Backblaze R2.

• AWS_S3_FORCE_PATH_STYLE - Use path style for the endpoint instead of the default subdomain style, useful for MinIO. Default false

• RUN_ON_STARTUP - Run a backup on startup of this application then proceed with making backups on the set schedule.

• BACKUP_FILE_PREFIX - Add a prefix to the file name.

• BUCKET_SUBFOLDER - Define a subfolder to place the backup files in.

• SINGLE_SHOT_MODE - Run a single backup on start and exit when completed. Useful with the platform's native CRON schedular.

• SUPPORT_OBJECT_LOCK - Enables support for buckets with object lock by providing an MD5 hash with the backup file.

• BACKUP_OPTIONS - Add any valid pg_dump option, supported pg_dump options can be found here. Example: --exclude-table=pattern