Chaos project, is a technical architecture to improve the resilience capacity of the complex technical means. Chaos after engineering experiments to ensure system availability. Chaos project aims to fault strangled in its infancy, that is, before the failure disruption will recognize them out. By actively manufacturing fault, test the system behavior under various pressures, failure to identify and fix problems and avoid serious consequences.
It was described as "conduct experiments on a distributed system of discipline in order to build confidence in a production environment to withstand turbulent conditions on the ability of the system.".
Chaos Engineering is the discipline of experimenting on a systemin order to build confidence in the system’s capabilityto withstand turbulent conditions in production.
It can also be seen as an influenza vaccine, deliberately injecting harmful substances in the body to prevent future disease, which seems crazy, but this method is also applicable to a distributed cloud system. Chaos project will fault injection system to a test system in response to it. This allows companies to prepare for the downtime, and downtime occurred before its impact to a minimum.
How do I know whether the system is in a steady state? Typically, a team may be verified through unit testing, integration testing and performance testing means. However, no matter how good these written tests, we believe that all is not enough, because errors can occur at any time, especially for distributed systems, then you need to introduce chaos Engineering (Chaos Engineering).
Failure Walkthrough: precipitation goal is common failure mode for online playback on controllable costs to sustained operational exercises and return way to expose the problem, continue to push forward systems, tools, processes, personnel capabilities.
