You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@storm.apache.org by GitBox <gi...@apache.org> on 2019/10/01 08:26:53 UTC

[GitHub] [storm] cjljohnson commented on a change in pull request #3131: STORM-3510: Track overflow count per taskId for resending backpressur…

cjljohnson commented on a change in pull request #3131: STORM-3510: Track overflow count per taskId for resending backpressur…
URL: https://github.com/apache/storm/pull/3131#discussion_r329934013
 
 

 ##########
 File path: storm-client/src/jvm/org/apache/storm/daemon/worker/WorkerState.java
 ##########
 @@ -88,6 +88,7 @@
 
     private static final Logger LOG = LoggerFactory.getLogger(WorkerState.class);
     private static final long LOAD_REFRESH_INTERVAL_MS = 5000L;
+    private static final int RESEND_BACKPRESSURE_SIZE = 10000;
 
 Review comment:
   It seems like such a niche situation that you would need to optimise this parameter to me.  It's only an issue if you run out of allocated memory before the backpressure signal gets through, so i'd think you'd need to be sending huge tuples which would fill your allocated memory before one of the status updates sent every 10000 tuples got through.  Maybe someone has a workload like that, but if so it's probably best for them to open a separate issue if it's an actual problem for them.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services