[
https://issues.apache.org/jira/browse/DIRMINA-331?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12464122
]
Julien Vermillard commented on DIRMINA-331:
-------------------------------------------
Thanks for this patch, I'm reviewing it now,
first comments :
easly fixable :
- code style is not the good one for 1.X branch, I know it's annoying :o)
- it's using java.util.concurrent in place of the backport
more problematic :
- it's preventing HTTP like protocols to calculate an average session
throughput
- doesn't give per service instantaneous throughtput stats, but total values
(service life wide total)
- adding ScheduledWrites and QueuedEvents is a good idea but doesn't give those
stats for the polling period, but again a total value
> StatCollector is not thread safe and some stats are being mixed up
> ------------------------------------------------------------------
>
> Key: DIRMINA-331
> URL: https://issues.apache.org/jira/browse/DIRMINA-331
> Project: MINA
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.0.1
> Reporter: Gaston Dombiak
> Fix For: 1.1
>
>
> The instance variable polledSessions is not thread-safe (i.e. uses ArrayList)
> and its access is not always protected (e.g. StatCollector$Worker#run()).
> Under heavy load this is producing an error in the Worker.
> The bytesReadThroughput statistic is incorrectly being set. It is adding up
> the throughput of written messages.
> Instance variables: msgWrittenThroughput, msgReadThroughput,
> bytesWrittenThroughputand bytesReadThroughput could be corrupted and return
> an unexpected value since they are not thread safe and 2 different threads
> may be changing them.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
