Idempotency Key
A client-supplied identifier used by a service to recognize retries of the same logical request and avoid duplicate side effects.
Distributed-Systems Context
An idempotency key identifies retries of the same logical request so a service can return the original result instead of repeating a side effect. Payment, record-creation, and external-API workflows need the key and operation outcome to be associated atomically, with explicit rules for key scope, retention, and payload mismatches.
Failure-Model Boundary
Storing a key in a cache is not sufficient if a crash can occur between the side effect and the key update. Transaction boundaries and recovery behavior must close that duplicate-execution window.
Related Distributed-Systems Concepts
- Idempotency
- Retry
- Transactional Outbox
- Atomicity