[jira] [Resolved] (FALCON-46) Flacon's instance retry service may hang because of an issue in InMemoryQueue

["Shaik Idris Ali (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FALCON-46?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Shaik Idris Ali resolved FALCON-46.
-----------------------------------

       Resolution: Fixed
    Fix Version/s: 0.3

Already backported from ivory:

https://git-wip-us.apache.org/repos/asf?p=incubator-falcon.git;a=commit;h=ce2589e2b36786a4fe3e61163ad58b62ccbe0e37

                
> Flacon's instance retry service may hang because of an issue in InMemoryQueue
> -----------------------------------------------------------------------------
>
>                 Key: FALCON-46
>                 URL: https://issues.apache.org/jira/browse/FALCON-46
>             Project: Falcon
>          Issue Type: Bug
>          Components: rerun
>    Affects Versions: 0.3
>         Environment: Oozie 3.2, CDH3U5 
>            Reporter: Shaik Idris Ali
>            Priority: Critical
>             Fix For: 0.3
>
>
> Falcon provides LateRerunService and RetryService to automatically handle late data reruns and instance failure correspondingly, for both process and a feed instance based on the policies like periodic, back-off etc.
> The service supports switchable queues to handle the messages, currently Retry Service uses InMemoryQueue which is built over Java's DelayQueue. A case where if the delay (delayedMessage) in message is greater than the existing message in the queue is not handle, and the service just stops enqueuing or dequeing the messages. Hence Falcon stops all the failed instance retries.
> A fix in the compareTo method of the event message is required. 
> The fix can be back-ported from Ivory.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira