[
https://issues.apache.org/jira/browse/NIFI-1155?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15003566#comment-15003566
]
ASF subversion and git services commented on NIFI-1155:
-------------------------------------------------------
Commit 3ed0949c5578f379cab7b90ff32778bf6296404a in nifi's branch
refs/heads/master from [~markap14]
[ https://git-wip-us.apache.org/repos/asf?p=nifi.git;h=3ed0949 ]
NIFI-1155: Ensure that when poll(FlowFileFilter, Set) is called, we properly
update the indicator for whether or not queue is full
Signed-off-by: joewitt <joew...@apache.org>
> Processor stops running if its destination applies backpressure and
> expiration occurs
> -------------------------------------------------------------------------------------
>
> Key: NIFI-1155
> URL: https://issues.apache.org/jira/browse/NIFI-1155
> Project: Apache NiFi
> Issue Type: Bug
> Components: Core Framework
> Affects Versions: 0.4.0
> Reporter: Mark Payne
> Assignee: Mark Payne
> Priority: Blocker
> Fix For: 0.4.0
>
> Attachments:
> 0001-NIFI-1155-Ensure-that-when-poll-FlowFileFilter-Set-i.patch,
> 0002-NIFI-1155-Refactored-StandardFlowFileQueue-to-update.patch
>
>
> If we setup a flow where Processor A is connected to Processor B and set
> backpressure to say 200 FlowFiles, all works well. If we then apply an ageoff
> (say 15 seconds) and start the Processors, all works well.
> Then stop Processor B so that the queue applies backpressure.
> Leave Processor B stopped long enough for the expiration to occur.
> The queue is now emptied, but Processor A does not continue to run.
> It appears that when FlowFiles are expired while the queue is full, the queue
> still thinks it is full.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
