[
https://issues.apache.org/jira/browse/STORM-2458?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sourav Mitra updated STORM-2458:
--------------------------------
Description:
Currently in the PartitionManager we just log the message and move on after
retryFurther() returns false.
There should be some method called on the FailedMsgRetryManager interface that
gives a chance to write CustomRetryManagers, to do something with that
offset/partition before just removing it. This might help STORM-1869 also.
Relevant PR: https://github.com/apache/storm/pull/2046
was:
Currently in the PartitionManager we just log the message and move on after
retryFurther() returns false.
There should be some method called on the FailedMsgRetryManager interface that
give a chance to write CustomRetryManagers, to do something with that
offset/partition before just removing it. This might help STORM-1869 also.
Relevant PR: https://github.com/apache/storm/pull/2046
> Kafka Spout should allow hooking up custom logic when retries have exhanusted
> -----------------------------------------------------------------------------
>
> Key: STORM-2458
> URL: https://issues.apache.org/jira/browse/STORM-2458
> Project: Apache Storm
> Issue Type: Improvement
> Components: storm-kafka
> Reporter: Sourav Mitra
> Priority: Minor
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Currently in the PartitionManager we just log the message and move on after
> retryFurther() returns false.
> There should be some method called on the FailedMsgRetryManager interface
> that gives a chance to write CustomRetryManagers, to do something with that
> offset/partition before just removing it. This might help STORM-1869 also.
> Relevant PR: https://github.com/apache/storm/pull/2046
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
