This session demonstrates how we enable self-healing from the root case of a failure.
A network failure in a complex system can have ripple effects throughout the system. For an application, especially in the world of NFV (but not only), quick recovery from failure is crucial.
In this presentation we will show how Heat uses Vitrage and Mistral to guarantee self-healing of the Heat stack - either on the physical or virtual layer. Vitrage, the OpenStack Root Cause Analysis service, is used to analyze the system state and identify the affected resources. Mistral, the OpenStack workflow service, provides a workflow for healing, and Heat automates the entire healing process.
We will demonstrate how these three projects work together to provide a new and easy way to self-heal your application with accuracy, and how to keep it unbroken and make sure everyone can use it. We will also introduce the self-healing SIG and discuss its future plans.
Attendees will learn about the self-healing SIG and its goals.
In the practical aspect, attendees will learn how to operate self-healing for instances and prevent a failure in the application in a simple manner. They will understand the power of Heat, Vitrage and Mistral, the strengths of each project and how they can benefit by combining them to a whole self-healing solution.
Attendees will also learn the details such as how to test the self-healing process and where the resources are.