Durable execution and state restoration is also a great way to get close to the crashing point again, though. The entire point of crashing early is to reset th

Durable execution and state restoration is also a great way to get close to the crashing point again, though. The entire point of crashing early is to reset the state of the program completely, hoping that the path that crashes it will not be taken too often. (and it only works well in the situations where crashing means negligible interruption of the service) By preserving the state between crashes, you diminish the pool of possible non-crashing states.