[
https://issues.apache.org/jira/browse/NIFI-2865?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15546630#comment-15546630
]
ASF GitHub Bot commented on NIFI-2865:
--------------------------------------
GitHub user markap14 opened a pull request:
https://github.com/apache/nifi/pull/1097
NIFI-2865: Refactored PublishKafka and PublishKafka_0_10 to allow bat…
…ching of FlowFiles within a single publish and to let messages timeout if
not acknowledged
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/markap14/nifi NIFI-2865
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/nifi/pull/1097.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #1097
----
commit 5b10a50ff34150c2642b544e4ee7a855c080b285
Author: Mark Payne <[email protected]>
Date: 2016-09-08T23:37:35Z
NIFI-2865: Refactored PublishKafka and PublishKafka_0_10 to allow batching
of FlowFiles within a single publish and to let messages timeout if not
acknowledged
----
> Address issues of PublishKafka blocking when having trouble communicating
> with Kafka broker and improve performance
> -------------------------------------------------------------------------------------------------------------------
>
> Key: NIFI-2865
> URL: https://issues.apache.org/jira/browse/NIFI-2865
> Project: Apache NiFi
> Issue Type: Bug
> Components: Extensions
> Reporter: Mark Payne
> Assignee: Mark Payne
> Fix For: 1.1.0
>
>
> When NiFi is unable to communicate properly with the Kafka broker, we see the
> NiFi threads occasionally block. This should be resolvable by calling the
> wakeup() method of the client. Additionally, if Kafka takes too long to
> respond, we should be able to route the FlowFile to failure and move on.
> PublishKafka has a nice feature that allows a demarcated stream to be sent as
> separate messages, so that a large number of messages can be sent as a single
> FlowFile. However, in the case of individual messages per FlowFile, the
> performance could be improved by batching together multiple FlowFiles per
> session
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
