Github user revans2 commented on the pull request:
https://github.com/apache/storm/pull/750#issuecomment-143259024
@HeartSaVioR I ran a number of tests, and I am fairly sure that what you
are seeing is a failure in the messaging layer and not disruptor. I limited
the number of workers to just 1 so that I would be sure to isolate things.
```storm jar storm_perf_test.jar com.yahoo.storm.perftest.Main --ack
--ackers 8 --bolt 8 --maxPending 200 --workers 1 --spout 8 --testTime 8000
--pollFreq 30```
I didn't run for the full 8000 seconds, but part way through I got this
screen shot.
I hope 185 million tuples from this test plus the 35 million tuples from my
in-order test all with no failures is enough to convince you that the new
disruptor queue is solid. If you want me to try and debug the messaging layer
and try to reproduce the failures you saw I am fine with that, but I would like
to do it as a part of another JIRA.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
