@tcrawley Thanks for that. In that case, is there any sort of mitigation measure for the pause during re-deploy when your old instance is gone and your new instance hasn't fully started yet?

@mingp: with uberjars, no. If you are running the app in a WildFly cluster, you can schedule jobs as singletons on every node. One node will be master for that job, but if it goes down (for a redeploy or whatever), another node will take over automatically.

@tcrawley I see. Thanks.