Skip to main content

ankra backup

Commands for managing the organisation’s backup infrastructure.

ankra backup restore-points

List the organisation’s restore points across every cluster

ankra backup restore-points list

List every restore point the organisation holds, newest first, across every cluster and stack. Examples:
Flags

ankra backup vaults

Manage the organisation’s backup vaults: S3-compatible object-storage targets that cluster backups are written to. The platform verifies each vault’s credentials against its bucket and reports the outcome as the vault’s status.

ankra backup vaults contents

List what a backup vault’s bucket actually holds, per restore point. This reads the object store, not Ankra’s rows, so it can disagree with the platform - which is the point. Use it to confirm that deleting a restore point removed its objects, and to find objects no restore point accounts for. Two prefixes are reported per restore point because they differ:
Volume data lives in a SHARED repository per cluster and namespace. Its bytes are reported separately and never counted into a single restore point: the blocks are deduplicated across every backup of that namespace, so no restore point owns them and deleting the prefix would destroy other restore points’ data. A vault whose bucket cannot be read fails with an error rather than printing an empty listing. An unreadable vault and an empty vault are different facts.
Flags

ankra backup vaults create

Create a backup vault: an S3-compatible bucket cluster backups are written to. The access keys are prompted for interactively when not passed as flags, so they never have to appear in your shell history. The platform verifies the keys against the bucket immediately and the command reports the outcome. Example:
Flags

ankra backup vaults delete

Delete a backup vault. By default this removes only Ankra’s record of the vault and the access keys it stored: the bucket, everything in it, and any provider resource Ankra created for it are left in your cloud account. --destroy-provider-resources also destroys what Ankra created for an Ankra-provisioned vault - it empties and deletes the bucket, and removes the UpCloud object storage service or DigitalOcean Spaces key that was minted for it. Restore points in that bucket are gone for good. It is refused for a vault that registers a bucket you created yourself.
Flags

ankra backup vaults get

Describe a backup vault: its endpoint, bucket, verification status, and - when the last credential check failed - the failure excerpt.
Flags

ankra backup vaults list

List the organisation’s backup vaults
Flags

ankra backup vaults provision

Create a backup vault and let Ankra create the bucket for it, using one of the organisation’s provider credentials (Hetzner, UpCloud, DigitalOcean or Scaleway). Ankra creates the bucket, mints or stores the access keys, verifies the bucket and registers the vault; the vault shows “provisioning” until that finishes. Everything is decided for you unless you say otherwise: the name defaults to “backups” (then “backups-2” and so on), the credential to the only one Ankra can provision from, and the region to that provider’s usual one. The command prints what it chose before it creates anything. Hetzner alone needs its Object Storage key pair passed in (or prompted for): Hetzner issues those in the Cloud Console (Object Storage > Manage credentials) and its Cloud API cannot mint them. The other providers need nothing beyond the credential. Examples:
Flags

ankra backup vaults verify

Re-run the platform’s credential check against the vault’s bucket and report the new status. Use this after rotating or fixing the access keys.