deployment-rollback

Deployment rollback automates recovery by reverting to a known-good container state using kubectl commands. The Skill detects deployment failures through health checks and smoke tests, then executes the rollback strategy, verifies the restored state, and notifies stakeholders of the incident and recovery.

Yes, rollback handles both containerized environments. The Skill assesses the deployment type, selects the appropriate strategy—container rollback via kubectl or docker-compose commands—executes the recovery, and runs post-rollback verification to confirm the known-good state is restored.

Database rollback is a distinct strategy executed during the phased rollback workflow. The Skill can revert database migrations using manage.py or equivalent tools, coordinate with container rollback, and verify data integrity before marking recovery complete.

The Skill implements automatic triggers for deployment failures, health check failures, and smoke-test failures. When detected, it assesses the failure, selects a rollback strategy, executes the recovery, and generates incident reports without manual intervention.

Yes, feature flag disable is a rollback strategy option. The Skill coordinates this with container and database rollback strategies, allowing you to mitigate failed features independently of full deployment reversion while maintaining service stability.

Post-rollback verification includes health checks and smoke tests to confirm the restored state is operational. The Skill reports verification results and stakeholder notifications, ensuring visibility into recovery success before marking the incident resolved.