|
|
Application errors occur more often than system failures. The first option to increase process or application availability is therefore to try a local restart of the failed application. If successful, this approach minimizes downtime as a process restart is significantly faster than a failover to a remote system. See 3.4, Restarting WebSphere processes as an OS service for information about how to do this.
In fact, there is no need for a highly available Deployment Manager in WebSphere V6. All critical singleton services can now run on different processes in the cell and are controlled by the HAManager. Therefore, the Deployment Manager is no longer a single point of failure. See Chapter 3, WebSphere administrative process failures for more information.
If you decide however, that you still want to make your Deployment Manager highly available, then the information provided in this section should be helpful.
