EclipseLink failover, connection pool, etc.

[William Cai-6]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi list,

I'm using EclipseLink JPA interface in my code and would confirm some issues with you people. Any response is greatly appreciated!

1. EclipseLink Connection Pool. According to the document, "For non-Java EE applications, you typically use internal connection pools. By default, EclipseLink sessions use internal connection pools.", and I do see multiple connections are started up when my application runs. However, in another page, I see "You can create internal connection pools only for server sessions (not for any other session type, including database sessions)", how to understand the statement?

2. Does EclipseLink provide any failover functionalities, for example specifying more than one JDBC URLs, etc? I did search the document and found nothing. Just want to confirm with the community to ensure I don't miss anything.

Again thanks for your help!

Thanks,
William

_______________________________________________
eclipselink-users mailing list
eclipselink-users@...
https://dev.eclipse.org/mailman/listinfo/eclipselink-users

[David Goodenough]

On Tuesday 23 June 2009, William Cai wrote:

> Hi list,
>
> I'm using EclipseLink JPA interface in my code and would confirm some
> issues with you people. Any response is greatly appreciated!
>
> 1. EclipseLink Connection Pool. According to the document, "For non-Java EE
> applications, you typically use internalconnection pools. By default,
> EclipseLink sessions use internal connection pools.", and I do see multiple
> connections are started up when my application runs. However, in another
> page, I see "You can create internal connection pools only for server
> sessions (not for any other session type, including database sessions)",
> how to understand the statement?
>
> 2. Does EclipseLink provide any failover functionalities, for example
> specifying more than one JDBC URLs, etc? I did search the document and
> found nothing. Just want to confirm with the community to ensure I don't
> miss anything.

You could try using HA-JDBC behind Eclipselink.  It does exactly this but
without changing the basic app (in this case EclipseLink).

David
>
> Again thanks for your help!
>
> Thanks,
> William


_______________________________________________
eclipselink-users mailing list
eclipselink-users@...
https://dev.eclipse.org/mailman/listinfo/eclipselink-users

[William Cai-6]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Thanks David. That's helpful.

---

From: David Goodenough <david.goodenough@...>
To: eclipselink-users@...
Sent: Tuesday, June 23, 2009 5:27:25 PM
Subject: Re: [eclipselink-users] EclipseLink failover, connection pool, etc.

On Tuesday 23 June 2009, William Cai wrote:

> Hi list,
>
> I'm using EclipseLink JPA interface in my code and would confirm some
> issues with you people. Any response is greatly appreciated!
>
> 1. EclipseLink Connection Pool. According to the document, "For non-Java EE
> applications, you typically use internalconnection pools. By default,
> EclipseLink sessions use internal connection pools.", and I do see multiple
> connections are started up when my application runs. However, in another
> page, I see "You can create internal connection pools only for server
> sessions (not for any other session type, including database sessions)",
> how to understand the statement?
>
> 2. Does EclipseLink provide any failover functionalities, for example
> specifying more than one JDBC URLs, etc? I did search the document and
> found nothing. Just want to confirm with the community to ensure I don't
> miss anything.

You could try using HA-JDBC behind Eclipselink.  It does exactly this but
without changing the basic app (in this case EclipseLink).

David
>
> Again thanks for your help!
>
> Thanks,
> William


_______________________________________________
eclipselink-users mailing list
eclipselink-users@...
https://dev.eclipse.org/mailman/listinfo/eclipselink-users

_______________________________________________
eclipselink-users mailing list
eclipselink-users@...
https://dev.eclipse.org/mailman/listinfo/eclipselink-users

[James Sutherland]

1 - An EclipseLink ServerSession is always used with JPA.  The doc is just saying the native API DatabaseSession a connection pool is not used, only a signle connection (but DatabaseSession is never used with JPA).

2 - EclipseLink does support failover with the Oracle database, and others should also be possible.  I think the failover database is included in the JDBC URL.  EclipseLink will also automatically ping, reconnect and retry failed queries, and a failover wait time can also be configured.  If your JDBC driver does not support failover, you could probably use EclipseLink SessionEvents to switch databases.

William Cai-6 wrote:

Hi list,

I'm using EclipseLink JPA interface in my code and would confirm some issues with you people. Any response is greatly appreciated!

1. EclipseLink Connection Pool. According to the document, "For non-Java EE applications, you typically use internalconnection pools. By default, EclipseLink sessions use internal connection pools.", and I do see multiple connections are started up when my application runs. However, in another page, I see "You can create internal connection pools only for server sessions (not for any other session type, including database sessions)", how to understand the statement?

2. Does EclipseLink provide any failover functionalities, for example specifying more than one JDBC URLs, etc? I did search the document and found nothing. Just want to confirm with the community to ensure I don't miss anything.

Again thanks for your help!

Thanks,
William