[jboss-jira] [JBoss JIRA] (WFCORE-2188) IO subsystem requires reload when increasing size of io-threads or task-max-threads

[Tomaz Cerar (JIRA)]

     [ https://issues.jboss.org/browse/WFCORE-2188?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Tomaz Cerar moved JBEAP-8251 to WFCORE-2188:
--------------------------------------------

              Project: WildFly Core  (was: JBoss Enterprise Application Platform)
                  Key: WFCORE-2188  (was: JBEAP-8251)
             Workflow: GIT Pull Request workflow   (was: CDW with loose statuses v1)
          Component/s: IO
                           (was: IO)
    Affects Version/s:     (was: 7.1.0.DR8)


> IO subsystem requires reload when increasing size of io-threads or task-max-threads
> -----------------------------------------------------------------------------------
>
>                 Key: WFCORE-2188
>                 URL: https://issues.jboss.org/browse/WFCORE-2188
>             Project: WildFly Core
>          Issue Type: Enhancement
>          Components: IO
>            Reporter: Tomaz Cerar
>            Assignee: Tomaz Cerar
>            Priority: Critical
>
> IO subsystem in EAP requires reload when increasing size of io-threads or task-max-threads.
> Open source competition like Payara doesn't require server reload when tuning thread pools. Example: https://www.youtube.com/watch?v=Nyq4vcEB6oc#t=9m21
> EAP 7 should at least allow increasing the size without reload, decreasing can be a bit complicated but worth exploring so to say.
> Customer use case: application in production, administrator detects increased load, wants to mitigate it a bit. Option would be to increase number of workers as the HW allows that, but can't do that as it requires reload and current clients would be in trouble.
> CLI example:
> {code}
> [standalone at localhost:9990 /] /subsystem=io/worker=default:write-attribute(name=io-threads, value=40)
> {
>     "outcome" => "success",
>     "response-headers" => {
>         "operation-requires-reload" => true,
>         "process-state" => "reload-required"
>     }
> }
> {code}
> {code}
> [standalone at localhost:9990 /] /subsystem=io/worker=default:write-attribute(name=task-max-threads, value=140)
> {
>     "outcome" => "success",
>     "response-headers" => {
>         "operation-requires-reload" => true,
>         "process-state" => "reload-required"
>     }
> }
> {code}



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