All the configuration for this matter can be found at Backup Configuration documentation. By default, backups are scheduled daily (config.backup.fullSchedule) at 05:00 UTC and with a retention policy (config.backup.retention) of 5 full-backups removed on rotation. You will have to find out the correct time window and retention policy that fit your needs.
In the next section, you’ll be able to see how to done this via Helm, with more explicit examples.
StackGres support Backups with the following storage options:
Examples are using MinIO service as a S3 compatible service for quick setups on local Kubernetes Cluster. Although, for production setups, StackGres Team recommends emphatically to pick a Storage as a Service for this purpose.
All the related configuration for the storage, is under configurations.backupconfig.storage section in your Stackgres Cluster configuration file.
configurations:
backupconfig:
# fill the preferred storage method with
# specific credentials and configurations
storage:
s3: {}
s3Compatible: {}
gcs: {}
azureBlob: {}
To extend the CRD for the backups, all the reference can be found at CRD Reference Documentation.
StackGres can perform a database restoration from a StackGres backup by just setting the UID of the backup CR that represents the backup that we want to restore. Like this:
cluster:
initialData:
restore:
fromBackup: #the backup UID to restore