How are rollbacks performed for faulty deployments?
amaise has an automated rollback mechanism:
Identification of the faulty change in the production environment
Rollback of the code change via the CI/CD pipeline — the previous stable version is automatically redeployed
Database: Restoration from the automatic pre-deployment snapshot if needed (approx. 30 minutes RTO)
Synchronization: Rollbacks are consistently applied across all environments
Thanks to the automatic pre-deployment database snapshots before each production release, a quick restoration of the database state is always possible. Deployments happen as rolling updates, so faulty containers are automatically stopped if health checks fail.