Right now a failure in machine provisioning (for example an out of quota failure) cause cloudify to re-try infinately a provisioning of the machine.
1. This lead to a potential error/leak as un-install attempt doesn't clear the machine.
2. in debug mode i want to know on those failure immediately and abort with a proper error rather than continuously retry and hide the error.
My suggestion for the debug option is that -disableSelfHealing will apply not just to failure of application failure but also to machine failure.
This implies a cloud driver 'debug' feature mode that will only allow a single failure, implemented above the cloud driver so that it applies to all cloud drivers.