[jboss-jira] [JBoss JIRA] (WFLY-8966) EJB subsystem thread-pool statistics largest-thread-count is counting threads instead of show the peak of parallel use

[Wolf-Dieter Fink (JIRA)]

    [ https://issues.jboss.org/browse/WFLY-8966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13428808#comment-13428808 ] 

Wolf-Dieter Fink commented on WFLY-8966:
----------------------------------------

A simple application is attached which use a simple scheduled EJB timer in a singleton (every 5sec)
So there is only one thread "simultaneously" running.
The expectation is that the pool only contains 1 thread as it is not used concurrent (and the default keepalive is 100millis).

But the largest-thread-count simple increases to 10 (max-threads) after the timer is started 10 times

> EJB subsystem thread-pool statistics largest-thread-count is counting threads instead of show the peak of parallel use
> ----------------------------------------------------------------------------------------------------------------------
>
>                 Key: WFLY-8966
>                 URL: https://issues.jboss.org/browse/WFLY-8966
>             Project: WildFly
>          Issue Type: Bug
>          Components: EJB
>    Affects Versions: 11.0.0.CR1
>            Reporter: Wolf-Dieter Fink
>         Attachments: ejb31-timer.jar
>
>
> The description of "largest-thread-count" is 
>  => "The largest number of threads that have ever simultaneously been in the pool."
> But the behaviour at runtime shows that it is handled similar to task-count.
> If only one thread is used it will increased for every invocation until max-threads.
> As result the largest-thread-count is == max-threads after a short time.
> The helpful metrict to check whether the threads are exhaused is missing!



--
This message was sent by Atlassian JIRA
(v7.2.3#72005)