03-18, 10:40–11:10 (Europe/Paris), VIP Area
Understanding what is happening in your cluster can be challenging. How can you quickly and easily tell if your cluster and apps are healthy, well utilized and running as expected?
In this tutorial, we'll look at various aspects of Kubernetes observability, and present multiple OSS solutions from the CNCF landscape and beyond to achieve that.
We will start with tools that simply query the Kubernetes API and deliver the output in an easy-to-understand UI (e.g. Skooner, k9s), go over sidecar-based and eBPF-based services meshes (e.g. Istio/Kiali, Cilium/Hubble UI) and end with application-side logging and monitoring (e.g. OpenTelemetry, fluentd, Jaeger, Grafana). Each level of observability demands a certain price in terms of configuration and runtime overhead. In turn the quality and depth of the information is different.
The intended take-away is to get an understanding which type of tooling is the right one for a given purpose. Most options will be shown in a live demonstration.
Out of the box, vanilla Kubernetes provides a few tools to help with observability such as basic logging with kubectl logs and basic metrics if you have metrics-server. This alone is not enough. We don't have fine-grained metrics, logs, data retention, or tracing, for example. These things can empower you when planning cluster capacity, troubleshooting bugs, or proactively anticipating operational issues, which are all steps in being production ready.
We want to help others decide which aspects and technologies should be considered when approaching observability of their clusters and apps.
We will show them one way to set up an observability stack using some of the many tools in the CNCF landscape as well as other OSS technologies that we have personal experience with. We will also talk about how mature these solutions are, what they can bring now and what to expect in the future.
Matthias Haeussler is Chief Technologist at Novatec Consulting, university lecturer for distributed systems, awarded ambassador of Cloud Foundry and the organizer of the Stuttgart Cloud Foundry Meetup. He advises and enables clients on their cloud-native journey, supports implementations and legacy migrations. Prior to that he was employed at IBM R&D Germany for more than 15 years. He has teaching experience from distributed systems and modern software architecture lectures at multiple universities in Stuttgart (DHBW, HSE, HfT). Besides that he is frequent speaker at various national and international conferences and meetups. (e.g. KubeCon, Devoxx, OSS Summit, Cloud Foundry Summit, Spring IO).
