saga-orchestration

Saga patterns manage distributed transactions by coordinating multi-service workflows and executing compensating actions for partial failures. This approach ensures atomicity and rollback without requiring distributed locks across microservices.

You implement compensating transactions within a saga orchestration to trigger rollback actions when a step fails. The coordinator manages state transitions and step timeouts to handle partial failures across long-running business processes.

Implementing saga orchestration requires an existing microservices architecture and a messaging infrastructure. You must define service boundaries, transaction requirements, and failure modes before deploying the saga coordinator.

Use saga patterns when managing long-running workflows that span minutes to hours across microservices. This approach is ideal when you need atomicity and compensation for complex business processes without relying on distributed locks.

Design a saga coordinator by specifying your transaction flow, service boundaries, and failure modes. The orchestration uses this input to manage state transitions and execute compensating actions to atomically roll back reservations if a booking step fails.