Retry strategy
The hera.workflows.retry_strategy module provides retry strategy functionality, along with necessary dependencies such as retry policy.
RetryPolicy
An enum that holds options for retry policy.
Source code in src/hera/workflows/retry_strategy.py
on_error
Retry steps that encounter Argo controller errors, or whose init or wait containers fail
on_failure
Retry steps whose main container is marked as failed in Kubernetes
RetryStrategy
RetryStrategy configures how an Argo job should retry.
Source code in src/hera/workflows/retry_strategy.py
affinity
affinity: Optional[RetryAffinity] = None
affinity dictates the affinity of the retried jobs
backoff
backoff dictates how long should a job wait for before retrying
expression
the expression field supports the expression of complex rules regarding retry behavior
limit
the hard numeric limit of how many times a jobs should retry
retry_policy
retry_policy: Optional[Union[str, RetryPolicy]] = None
the policy dictates, at a high level, under what conditions should a job retry
build
build() -> RetryStrategy
Builds the generated RetryStrategy representation of the retry strategy.