[Pacemaker] Startup ordering problem

[Brett Delle Grazie]

Hi Jody,

On Fri, 2010-09-10 at 20:33 +0000, Jody Nickel wrote:
> Brett Delle Grazie <brett.dellegrazie at ...> writes:
> > 
> > Hi,
> > 
> > Comments interleaved below,
> > 
> > > 
> > > 
> > > 
> > 
> Brette,
>   Thanks for your response and I might end up trying something along those
>   lines but it gets even more complicated with Hibernate in the mix. The
>   other problem is how can I tell from within Tomcat whether there is a
>   real database problem or that the database has migrated to another node.
> 
Are your developers using Spring? That makes working with Hibernate
significantly easier.  Try to get the developers to use JNDI access to
the database pool rather than defining the pool within the hibernate
configuration (if they aren't already) as that provides you with
significantly greater control and better pooling (the container, i.e.
Tomcat, controls the pool rather than the application).

>   I had hoped to use the cluster to start/stop services so that I don't 
>   need to worry about trying to distinguish what type of database problem
>   has occurred from within the application.

As far as I can tell from your config, the start/stop ordering will only
apply to the node(s) running those items.  That is, Tomcat and Apache
will only be restarted if they are running on the Database nodes.

> 
>   I can't seem to get the ordering right with the configuration I've
>   included, the services stop/start but not with the ordering or restarting
>   that I thought pacemaker should enforce via the rules.
> 
> Jody
>   
> 
> 
> 
> 
> 

-- 
Best Regards,

Brett Delle Grazie

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________