Scenarios

The Examples page is a ladder of component manifests, one CRD capability per file. Scenarios are the layer above. Each one walks a real situation end to end, like “the cluster is gone, get my data back”, and ties together the resources you apply, the kubectl commands you run, and the checks that tell you it worked.

Each scenario is backed by one apply-ready bundle under deploy/examples/scenarios/. Copy it, replace the REPLACE_ME values, and kubectl apply -f.

The mental model (read this first if you’re new)

Kopiur splits one job into separate resources so each can change independently:

• a Repository is where snapshots live: an S3 bucket, a NAS, B2, and so on;
• a SnapshotPolicy is the recipe, meaning what to back up. It runs nothing on its own;
• a Snapshot is one invocation: a single snapshot as a Kubernetes object;
• a SnapshotSchedule is the cron, meaning when the recipe runs;
• a Restore reads a snapshot back into a PVC;
• a Maintenance reclaims space in the repository.

Identity is the detail that makes or breaks the recovery and migration scenarios. kopia stores each snapshot under username@hostname:path, defaulting to <policy-name>@<namespace>:/pvc/<pvcName>. To find an existing snapshot, you have to match that identity. See How Kopia works.

#	Scenario	When you reach for it	Complexity
01	Protect a stateful app	Nightly, application-consistent backups of a database on a PVC.	Simple
02	Recover from accidental data loss	Someone deleted/corrupted data; restore the last good snapshot safely.	Simple
03	Disaster recovery on a fresh cluster	The cluster is gone; rebuild the app + its data from the repository.	Advanced
04	Migrate an app across clusters / namespaces	Move a stateful app to a new namespace or cluster, data and all.	Advanced
05	Adopt an existing kopia repository	You already have a kopia repo; take it over without re-uploading.	Advanced
06	Backup verification / restore drills	Prove backups are restorable on a schedule, and alert when one isn’t.	Advanced
07	Point-in-time rollback	Roll a volume back to a specific known-good moment (asOf/offset).	Simple
08	Clone an app into another namespace	Copy prod data into staging/preview to reproduce a bug or seed an env.	Advanced
09	Share one repository across clusters	Several clusters back up to (or restore from) the SAME repository at once, safely.	Advanced
10	DR from a replicated repository	The primary repository is gone; seed a NEW one from the surviving off-site mirror (spec.seed).	Advanced

Alpha

These use API group kopiur.home-operations.com, version v1alpha1. Backends are externally tagged: the bucket lives under backend.s3, not backend: { kind: S3 }.

See also

• Getting started: the install-to-first-restore walkthrough.
• Examples: the per-capability manifest ladder.
• Backups & schedules and Restores: the field-level reference for the resources these scenarios combine.
• Troubleshooting: when a step doesn’t go green.