On the flip side I've seen dev teams grind to a halt because of how much difficulty they had tracing a request end to end in their system. Granted, they had en

On the flip side I've seen dev teams grind to a halt because of how much difficulty they had tracing a request end to end in their system. Granted, they had entirely too many separate web services, and that was really the root of their problem, but being able to trace a request all the way through your system isn't useful until it is (meaning, when shit hits the fan).