I have 2 applications which are deployed on two different clusters. Each cluster has got two server members. The first member is the active one and the second one is the failover member. So there s no loadbalancing per-se just failver over.
Now, I want to have a so called sorry server which may act as a third member to each cluster. The sorry server is suposed to work in the following fashion and given circumstances.
If I have a donwntime ( e.g. deploying a new app on the cluster ) and if someone tries to access the application. Now the first two members are down for downtime. I want the sorry server to serve the sorry pages for all the requests coming during the downtime.
I can be pretty flexible for the implementation. For some reason, hardware implementation is not possible.
I am using IBM IHS server in front of app-server cluster.
I can think of running two ihs server : one at a given time.
One ihs server would server request to usual cluster and other would server request to sorry-cluster. So, during downtime, I can shutdown the main ihs server and bring up the so called ihs-sorry server so thAT it can forward all requests to sorry-cluster
IBM Certified Advance WAS Administrator on 6.1, BEA Certified WebLogic Server 81 Administrator, Vitria Certified BusinessWorks Administrator, SCJP 2
I’ve looked at a lot of different solutions, and in my humble opinion Aspose is the way to go. Here’s the link: http://aspose.com