Recovery Testing. How and why to perform it?

Everyone is familiar with such types of load testing types as maximum performance and stability testing, as well as the tools used for them like JMeter, Gatling, and k6. However, to fully understand a system’s quality, it is important not just to know the system’s usage capacity, but also  to validate how well it recovers after unanticipated situations.

In this talk, we will focus on recovery testing—a key process that evaluates a system’s ability to return to normal behavior after crashes. We will discuss tools for this, including stress-ng and iptables.

We will cover the main steps of preparing and performing the tests, including setting up a test environment and simulating various crash scenarios.

In addition, we will discuss supplementary tools and approaches that are useful in the test process. This will help you better understand the importance of recovery testing and prepare the system for possible crashes.