[DBCP] Connection Pool Behavior When Database Goes Down

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

[DBCP] Connection Pool Behavior When Database Goes Down

Our application employs a somewhat modified/wrapped version of DBCP.
When the database goes down, or a the application cannot connect, our
application log file gets very large very quickly because the
application continues to request connections.

The good news is that when the connection again become available, the
app sever and the connection pool recovers nicely and starts chugging
along again.

Has anyone else experienced this behavior?  I looked at the
configuration settings for DBCP, and I didn't see any settings that
might ameliorate the problem.  

I guess the general topic is DBCP behavior when the database becomes
unavailable for some undetermined period of time. Does the intelligence
need to be put into the application, or can the connection pool help
manage this?

Does anyone know how WebLogic and WebSphere's datasource connection
pools handle this scenario?

To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]