As more OpenStack clouds enter production, operational management is becoming a bigger focus for customers. Enterprises have varying requirements, and there are a myriad of tools out there to meet those requirements. In this talk, we will outline how a developer or component-focused approach is not what customers want, and we will provide guidance on performance and capacity, sizing, and availability monitoring. We will also suggest approaches for using tools like Sensu, Fluentd, and Collectd, including how they can be integrated with OpenStack deployments. In addition, we will discuss templates for what should be monitored for OpenStack and how to prepare a manifest so that OpenStack operational management tools can be easily integrated into Enterprise customer operations practices.
- What to monitor in production OpenStack clouds
- How to integrate tools such as fluentd, collectd and sensu in production OpenStack clouds
- How to build and provide manifests of operational management actions provided by OpenStack deployments to integrate with enterprise customer operations processes