[jboss-jira] [JBoss JIRA] (AS7-3848) @Clustered @Stateless SLSB should send topology information to the client

jaikiran pai (JIRA) jira-events at lists.jboss.org
Wed Feb 29 09:58:37 EST 2012


    [ https://issues.jboss.org/browse/AS7-3848?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12672152#comment-12672152 ] 

jaikiran pai commented on AS7-3848:
-----------------------------------

To add more to this - we currently consider @Clustered @Stateless just as any other @Stateless. The EJB client API has the ability to route invocations (based on configurable algorithm) to different nodes within a cluster if it knows the cluster topology of the bean. If the topology isn't available then the invocations for the stateless bean always end up on the same server, which is constrained by the size of the bean instance pool used by the stateless bean. 

>From an implementation point of view, just like for stateful beans, the server side would have to do the following for stateless beans:

1) Send back a cluster topology to clients, when a client connects to a server and the server has @Clustered @Stateless beans deployed.
2) Send back cluster view change notifications
3) Send back weak affinity association for invocations even on stateless beans

                
> @Clustered @Stateless SLSB should send topology information to the client
> -------------------------------------------------------------------------
>
>                 Key: AS7-3848
>                 URL: https://issues.jboss.org/browse/AS7-3848
>             Project: Application Server 7
>          Issue Type: Feature Request
>          Components: Clustering, EJB
>    Affects Versions: 7.1.0.Final
>            Reporter: Radoslav Husar
>            Assignee: Paul Ferraro
>              Labels: eap6_LA, failover_testing
>             Fix For: 7.1.2.Final
>
>
> Having AS7-2887 fixed, we currently take no action on @Clustered @Stateless but we allow deployment.
> The remote ejb client then treats the node as standalone nodes, clustering subsystem is not started. This however means that the topology is static -- compared to AS5 client behavior.
> There are cases where it's useful mainly because the server side is constrained by the pool size of the SLSB (--Jaikiran).

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        


More information about the jboss-jira mailing list