if an application is deployed to the default context-root (/), what are some ways to implement a "maintenance mode" for the app.
I would like to display a "site down for maintenance" page while new versions are re-deployed, but with no downtime on the app.
the first idea that comes to mind is:
1) undeploy app with context-root /
2) deploy simple war with maintenance page to context-root /
but during deployment / undeployment, there will be a few moments where the application will not be available, and would like to have a seamless approach to this.
Another idea for a clustered environment:
shut down 1 node, start up on different partition, undeploy app, deploy maintenance app.
repeat process across all nodes in the cluster
during this all existing sessions will be maintained, but new requests will get directed to maintenance nodes.
ideas on this? the whole restart procedure is not ideal....