[
https://issues.apache.org/jira/browse/FLUME-1378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13417017#comment-13417017
]
Denny Ye commented on FLUME-1378:
---------------------------------
OK, it can be fixed with parameter checking. The impact of this bug is
terrible, infinite lifecycle of Sink, and huge heap with full GC in 90% time,
causes the 'GC overhead limit exceeded'
> Sink falls into infinite lifecycle with EventDeliveryException
> --------------------------------------------------------------
>
> Key: FLUME-1378
> URL: https://issues.apache.org/jira/browse/FLUME-1378
> Project: Flume
> Issue Type: Bug
> Components: Channel, Sinks+Sources
> Affects Versions: v1.2.0
> Reporter: Denny Ye
> Assignee: Denny Ye
> Fix For: v1.3.0
>
>
> I tested the Flume and found a critical issue. HDFSEventSink with
> 'txnEventMax' parameter > MemoryChannel 'transactionCapacity' parameter, Sink
> falls into infinite lifecycle and all Events has been hold in MemoryChannel
> with error message:
> "Take list for MemoryTransaction, capacity <n> full, consider committing more
> frequently,increasing capacity, or increasing thread count"
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
