Sometimes, a single component may not meet the requirement for your application. For example, in order to deploy a WordPress, you will need a Deployment for the WordPress and another Deployment for database to store it’s contents. Now, you may want to backup both of the deployment and database under a single configuration as they are parts of a single application.
Stash 0.9.0+ supports taking backup multiple co-related components using a single configuration known as BackupBatch. This guide will give you an overview how batch backup works in Stash.
The following diagram shows how Stash takes backup of multiple co-related components in a single application. Open the image in a new tab to see the enlarged version.
The backup process consists of the following steps:
At first, a user creates a backend Secret. This secret holds the credentials to access the backend where the backed up data will be stored.
Then, she creates a Repository
crd which represents the original repository in the backend.
Then, she creates a BackupBatch
crd which specifies multiple targets(workload, volume, and database). It also specifies the Repository
object that holds the backend information where the backed up data will be stored.
Stash operator watches for BackupBatch
objects.
When it finds a BackupBatch
object, it checks if there is any workload as a target. If there any, it injects a sidecar named stash
into the workloads.
It also creates a CronJob
to trigger backups periodically.
TheCronJob
triggers backup on each scheduled slot by creating a BackupSession
crd.
The BackupSession controller (inside sidecar for sidecar model or inside the operator itself for job model) watches for BackupSession
crd.
When it finds a BackupSession
it starts the backup process immediately(for job model a job is created for taking backup) for the individual targets.
The individual targets complete their backup process independently and update their respective fields in BackupSession
status.