Summary
As application developer
I want to have a very simple way to configure the backup
So that i can operate my application very fast
Context
With component-solution-backup, we have a way to enable backups for a namespace by adding two labels to the namespace.
Currently, Espejote does not support generating secrets, so we cannot eliminate the need for the second label. As a result, we still need to manually create the encryption secret.
If we integrate this feature into K8up itself, K8up could generate the encryption secret (though the user would need to back up the encryption secret manually). It would be beneficial if K8up could create an ObjectBucket per namespace to ensure isolation.
However, this may not be feasible on all clusters. For clusters that support provisioning ObjectBuckets via CRD, the cluster administrator could provide a template specifying how the ObjectBucket object should be created.
Additionally, implementing
would simplify this feature.
Out of Scope
No response
Further links
No response
Acceptance Criteria
- Given I enabled the feature and I label my namespace with k8up.io/backup: true, an ObjectBucket, a Schedule and an encryption secret are created and the PVC in the namespace are backed up.
Implementation Ideas
No response
Summary
As application developer
I want to have a very simple way to configure the backup
So that i can operate my application very fast
Context
With component-solution-backup, we have a way to enable backups for a namespace by adding two labels to the namespace.
Currently, Espejote does not support generating secrets, so we cannot eliminate the need for the second label. As a result, we still need to manually create the encryption secret.
If we integrate this feature into K8up itself, K8up could generate the encryption secret (though the user would need to back up the encryption secret manually). It would be beneficial if K8up could create an ObjectBucket per namespace to ensure isolation.
However, this may not be feasible on all clusters. For clusters that support provisioning ObjectBuckets via CRD, the cluster administrator could provide a template specifying how the ObjectBucket object should be created.
Additionally, implementing
spec.backend.s3.endpointandspec.backend.s3.bucketvalues to be taken from secrets #828would simplify this feature.
Out of Scope
No response
Further links
No response
Acceptance Criteria
Implementation Ideas
No response