[
https://issues.apache.org/jira/browse/NIFI-1237?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Bryan Bende updated NIFI-1237:
------------------------------
Attachment: NIFI-1237.patch
I was able to reproduce this by creating another test that executed
testBatching() a few hundred times in a loop.
The batching test was sending 20 messages and the processor has an internal
queue with capacity 10. Since we are sending all messages before triggering the
processor, the queue will fill up and the 10 other messages would need to be
held in the socket buffer.
The attached patch reduces it to 5 messages which is still sufficient for
testing batching and avoids dealing with the queue being full. Ran this several
thousand times in a loop and did not fail.
> TestListenSyslog has a race condition?
> --------------------------------------
>
> Key: NIFI-1237
> URL: https://issues.apache.org/jira/browse/NIFI-1237
> Project: Apache NiFi
> Issue Type: Bug
> Components: Extensions
> Reporter: Joseph Witt
> Assignee: Bryan Bende
> Fix For: 0.4.0
>
> Attachments: NIFI-1237.patch
>
>
> causes build to fail at times
> TestListenSyslog.testBatching:254 expected:<20> but was:<10>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
