[
https://issues.apache.org/jira/browse/METRON-1888?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16700712#comment-16700712
]
ASF GitHub Bot commented on METRON-1888:
----------------------------------------
Github user justinleet commented on the issue:
https://github.com/apache/metron/pull/1276
+1 by inspection. I'm okay with leaving this undefined by default.
Should we have a jira for the ability to detect (and maybe in a perfect
world, correct/mitigate) the max spout pending limit issue you mention above?
> Default Topology Settings in MPack Cause Profiler to Stall
> ----------------------------------------------------------
>
> Key: METRON-1888
> URL: https://issues.apache.org/jira/browse/METRON-1888
> Project: Metron
> Issue Type: Bug
> Reporter: Nick Allen
> Assignee: Nick Allen
> Priority: Major
>
> The default max spout pending set by the Mpack for the Profiler topology is
> 300. This value is too low and causes the topology to stall and stop
> consuming messages.
>
> The Profiler topology is the only Metron topology that uses Storm's windowing
> support. Storm will queue up messages for the window length (by default 30
> seconds) and process each of those windows.
>
> This causes Storm to need to "hang on" to tuples longer than the other
> topologies and this requires a much greater max spout pending value.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
