[jboss-jira] [JBoss JIRA] Commented: (JBAS-5253) RetryInterceptor that caches JNDI lookup details per EJB type

[Galder Zamarreno (JIRA)]

    [ http://jira.jboss.com/jira/browse/JBAS-5253?page=comments#action_12400578 ] 
            
Galder Zamarreno commented on JBAS-5253:
----------------------------------------

Possible workarounds:

1.- Deploy all remotely called EJBs homogeneously in a separate 
cluster and do HA-JNDI lookups passing the partition name as indicated 
in: http://wiki.jboss.org/wiki/Wiki.jsp?page=JBossHAJNDIUseCluster 

This workaround removes the need to use different HA-JNDI lookup 
details per EJB.

2.- Before each EJB remote call, call new InitialContext() passing 
the lookup details for that specific EJB. This will force the RI to update 
the JNDI lookup details with the ones from new InitialContext(). This 
workaround carries the performance penalty of having to do a new 
lookup for each remote call.

3.- Cache different remote EJB JNDI lookup details and before each 
remote call, assign them to the RI so that in case of complete cluster 
shutdown, it has got the right lookup details. This workaround gets around 
the need do a new lookup for every remote call. See 
http://wiki.jboss.org/wiki/Wiki.jsp?page=RetryInterceptor wiki to find an 
example on how to do this.

> RetryInterceptor that caches JNDI lookup details per EJB type
> -------------------------------------------------------------
>
>                 Key: JBAS-5253
>                 URL: http://jira.jboss.com/jira/browse/JBAS-5253
>             Project: JBoss Application Server
>          Issue Type: Feature Request
>      Security Level: Public(Everyone can see) 
>          Components: Clustering, Naming
>            Reporter: Galder Zamarreno
>         Assigned To: Galder Zamarreno
>            Priority: Minor
>
> Configuring a remote EJB client, whether standalone or within AS, to use 
> RetryInterceptor (RI) when EJBs are deployed across different node/clusters
> can be a bit tricky cos jndi details to be used by interceptor are stored statically. 
> There're several available workarounds which I'll explain in the workaround 
> section. This JIRA is focused at the feasibility of creating a more intelligent RI 
> that caches JNDI lookups as per EJB type.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira