Package org.apache.hadoop.io.retry

Interface RetryPolicy

All Known Implementing Classes:

RetryPolicies.MultipleLinearRandomRetry

@Evolving
public interface RetryPolicy

Specifies a policy for retrying method failures. Implementations of this interface should be immutable.

Nested Class Summary

Nested Classes

Modifier and Type	Interface	Description
static class	RetryPolicy.RetryAction	Returned by shouldRetry(Exception, int, int, boolean).

Method Summary

Modifier and Type	Method	Description
RetryPolicy.RetryAction	shouldRetry(java.lang.Exception e, int retries, int failovers, boolean isIdempotentOrAtMostOnce)	Determines whether the framework should retry a method for the given exception, and the number of retries that have been made for that operation so far.

Method Detail

shouldRetry

RetryPolicy.RetryAction shouldRetry(java.lang.Exception e,
                                    int retries,
                                    int failovers,
                                    boolean isIdempotentOrAtMostOnce)
                             throws java.lang.Exception

Determines whether the framework should retry a method for the given exception, and the number of retries that have been made for that operation so far.

Parameters:

e - The exception that caused the method to fail

retries - The number of times the method has been retried

failovers - The number of times the method has failed over to a different backend implementation

isIdempotentOrAtMostOnce - true if the method is Idempotent or AtMostOnce and so can reasonably be retried on failover when we don't know if the previous attempt reached the server or not

Returns:

RetryPolicy.RetryAction with RetryDecision.FAIL if the method should not be retried, RetryDecision.RETRY if the method should be retried or RetryDecision.FAILOVER_AND_RETRY if failover has to be performed before retry.

Throws:

java.lang.Exception - The re-thrown exception e indicating that the method failed and should not be retried further