[jira] [Resolved] (AMQ-8039) Allow Inactivity monitor thread pool configuration

["Gary Tully (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/AMQ-8039?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Gary Tully resolved AMQ-8039.
-----------------------------
    Resolution: Fixed

The following system properties allow the inactivity monitor thread pool executor to be configured. No change to current unbounded defaults

org.apache.activemq.transport.AbstractInactivityMonitor.keepAliveTime  30
org.apache.activemq.transport.AbstractInactivityMonitor.corePoolSize 0
org.apache.activemq.transport.AbstractInactivityMonitor.maximumPoolSize Integer.MAX_VALUE
org.apache.activemq.transport.AbstractInactivityMonitor.workQueueCapacity 0
org.apache.activemq.transport.AbstractInactivityMonitor.rejectWork false

> Allow Inactivity monitor thread pool configuration
> --------------------------------------------------
>
>                 Key: AMQ-8039
>                 URL: https://issues.apache.org/jira/browse/AMQ-8039
>             Project: ActiveMQ
>          Issue Type: Improvement
>          Components: Transport
>    Affects Versions: 5.16.0
>            Reporter: Gary Tully
>            Assignee: Gary Tully
>            Priority: Major
>             Fix For: 5.17.0
>
>
> In cases where there are lots of connections that timeout, there will be a one to one relation with the timeout task that stops the connection, this in turn fires an executor to close the underlying transport. 
> If any event of the stop blocks for a period there can be a large spike in inactivity monitor threads. Allowing inactivity monitor work to queue up in this case is sensible to place a bound on the number of threads in a broker process.
>  
> related: AMQ-4156 and AMQ-5480 AMQ-5486 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)