I found [this post](https://devtron.ai/blog/creating-a-kubernetes-cron-job-to-backup-postgres-db/) to be a good starting point for automating backups with [[PostgreSQL]] deployed within [[Kubernetes]]. The general concept is: 1. Create an [[S3]] bucket 2. Create a [[Docker]] container image which contains the PostgreSQL client and the [[AWS CLI]] 3. Create a [[cron]] job which will run a backup script periodically 4. The script will run pgsql dump, tarball up the result, and upload it to the S3 bucket However, there is a bug in that post's backup script. The `pg_dump` result should end in `.sql` rather than `.tar`. It doesn't result in any data loss but it's clearer.