Backup & Restore
The MSSQL Provider backs up SQL Server natively through the mssql-native
backup class. Backups run through SQL Server's BACKUP ... TO URL against
S3-compatible object storage, driven by the operator's MSSQLBackup,
MSSQLBackupSchedule, and MSSQLRestore resources — there is no side-car Job.
The mssql-native backup class
| Property | Value |
|---|---|
| Execution | Provider-managed (operator-native, no Job) |
| Storage | S3-compatible object storage |
| Point-in-time recovery (PITR) | Supported (standalone instances) |
| Requires | SQL Server 2022 or later |
Point-in-time recovery
SQL Server supports PITR by replaying the transaction-log chain
(full → differential → log) up to a target time. Enable it per storage with
spec.backup.storages[].pitr.enabled=true; the provider then archives
transaction-log backups on a schedule so a recovery chain always exists.
spec:
backup:
storages:
- name: s3-primary
pitr:
enabled: true
PITR restore into an Availability Group is not supported by the operator yet — only standalone instances.
How it maps to the operator
The provider translates platform Backup / Restore resources into the
operator's native backup CRDs. For the underlying mechanics — backup types,
retention, S3 target configuration, and restore semantics — see the operator's
Backups & Restores guide.
See also
- Operator: Backups & Restores.
- Versions — PITR requires SQL Server 2022+.