You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@synapse.apache.org by "Andreas Veithen (JIRA)" <ji...@apache.org> on 2008/08/03 14:21:44 UTC

[jira] Commented: (SYNAPSE-416) Improvements to the DataSource Connection Pool

    [ https://issues.apache.org/jira/browse/SYNAPSE-416?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619323#action_12619323 ] 

Andreas Veithen commented on SYNAPSE-416:
-----------------------------------------

I think that in order to keep things simple, the DB mediators should use a single mechanism to lookup data sources, namely JNDI as they do currently. Depending on the use case, we indeed don't want to expose resources registered in JNDI to other applications. The problem here is not JNDI itself, but the JNDI provider that is being used: the RMI registry provider. The solution is to use a pure in-memory JNDI provider. I did that successfully with Tomcat's naming provider.

The other problem is that currently it is required to configure the initial context factory on every instance of the DB mediators. In order to solve this we should have a mechanism that allows to configure a default provider and the required credentials such that in the configuration of the DB mediator, only the JNDI name is mandatory.

> Improvements to the DataSource Connection Pool
> ----------------------------------------------
>
>                 Key: SYNAPSE-416
>                 URL: https://issues.apache.org/jira/browse/SYNAPSE-416
>             Project: Synapse
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: NIGHTLY
>            Reporter: indika priyantha kumara
>            Assignee: indika priyantha kumara
>             Fix For: FUTURE
>
>
> Improvements to the DataSource Connection Pool
> 1) If the DB connection pools are defined in the synapse.properties, then , those need to be accessed just providing data source name. Simply, we have to differentiates the accessing datasources that we have created and datasources that we are looking up.
>            Then , For access datasources we have defined (Example dblookup mediator)
>                     <pool>
>                         <dsName>lookupds</dsName>                      
>                     </pool>
> 	and for any external datasources as before 
>                   <pool>
>                         <dsName>lookupds</dsName>
>                         <icClass>com.sun.jndi.rmi.registry.RegistryContextFactory</icClass>
>                         <url>rmi://localhost:2199</url>
>                         <user>esb</user>
>                         <password>esb</password>
>                     </pool>
>       11)  If the  datasources that we have defined is not intended to use by any other application. Then, we never want to register those in the JNDI tree. Those can be in a local pool. Currently, all are registered in the JNDI. This need to be configurable so that if someone want to avoid JNDI 	registration , he can tell it by configuration option.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@synapse.apache.org
For additional commands, e-mail: dev-help@synapse.apache.org