[
https://issues.apache.org/jira/browse/NIFI-1176?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15011241#comment-15011241
]
ASF subversion and git services commented on NIFI-1176:
-------------------------------------------------------
Commit 180ea1ba222a5890541dfbe046f2876695e69239 in nifi's branch
refs/heads/master from [~markap14]
[ https://git-wip-us.apache.org/repos/asf?p=nifi.git;h=180ea1b ]
NIFI-1176: Use a smaller internal blocking queue size of only 5000 messages; do
not create a new queue every time the processor is scheduled to run
Signed-off-by: Aldrin Piri <[email protected]>
> GetTwitter can cause OOME with small heaps if backpressure applied
> ------------------------------------------------------------------
>
> Key: NIFI-1176
> URL: https://issues.apache.org/jira/browse/NIFI-1176
> Project: Apache NiFi
> Issue Type: Bug
> Components: Extensions
> Reporter: Mark Payne
> Assignee: Mark Payne
> Fix For: 0.4.0
>
> Attachments:
> 0001-NIFI-1176-Use-a-smaller-internal-blocking-queue-size.patch
>
>
> GetTwitter queues up a lot of data in the background and then transfers the
> data via FlowFile when triggered to run. However, if backpressure is applied
> the queue can grow to the point of 100,000 tweets. We need to bring this down
> to a couple thousand max.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
