Rather, means deployed prior to the failed source remain in position, and the next deployment starts off ????? Together with the failed source. You will shell out a whole lot less time expecting deployments along with a ton much more time producing your infrastructure.