Software Architect's Handbook

Retry strategy

Transient faults are errors that occur due to some temporary conditions, such as network connectivity issues, temporary unavailability of a service, timeout of a service, or infrastructure-level faults.

Retry strategies can be used to attempt to retry an operation when it encounters a transient fault. A retry policy can be tailored based on the nature of the component that caused the error, which can dictate things such as how many retry attempts are made, and the length of time to wait in between attempts.

Some common retry interval types include:

• Regular intervals: The software system waits the same amount of time in between each interval attempt.
• Incremental intervals: The software system waits a short amount of time before ...