AWS doesn’t have a PaaS service offering for MongoDB at the moment. So it is the DevOps’s responsibility to maintain their own backups of the production MongoDB databases. So I’ve developed a custom script that makes use of Docker containers to deploy a backup container.
This script will generate backup in specified intervals and upload that to AWS S3 bucket. The AWS S3 buckets can be configured to have versioning as well as life-cycle management such as delete backups that are older than x days etc.
The project has been adjusted to make it open source and uploaded to github.
2 thoughts on “Automated MongoDB backup on Docker Swarm and save on AWS S3”
Why not just use Ops Manager?
You can. Our situation was different as we have already had several docker clusters and everything was managed/running as a docker container so it was the quickest way than deploying and managing another service.