Summary

This chapter has provided a basic understanding of how to write tooling that provides safety in the face of chaos. We have shown you how circuit breakers or exponential backoff can save your network and services from overload when unexpected problems occur. We have shown how rate-limiting automation can prevent runaway workflows before responders can react. You have learned about how tool scoping via a centralized policy engine can provide a second layer of safety without overburdening your developers. We have learned the importance of idempotent workflows to allow workflow recovery. And finally, we have conveyed how an emergency-stop system can be utilized by first responders by quickly limit damage to automation systems while investigating a problem.

At this time, if you haven't played with the workflow system that we have been developing, you should explore the code and play with the examples. The README.md file will help you get started. You can find this at the...