When: Thursday, March 30, 14:45 - 17:00
Where: C04, Berlin Congress Center (BCC)
This workshop, hosted by Alexandru Somesan and Frederic Branczyk from CoreOS, is designed for newcomers to the Kubernetes ecosystem that are looking to learn how to monitor their applications in the cluster using Prometheus.
Throughout this session, participants will learn how to reliably deploy Prometheus on Kubernetes, instrument and monitor own applications using the same principles by which Prometheus monitors the Kubernetes cluster itself and finally make use of collected metrics to set-up alarms and dashboards that suit their use-case.
- Every participant has a working K8S cluster (either BYO or ad-hoc local multinode)
- Deploy Prometheus on the cluster using the Operator
- Deploy example application (ideally web server + backend) to K8S cluster
- Discuss data flow from client to application in the context of K8S (ingress / external service)
- Discuss what kind of application metrics would meaningfully describe its operational state (counting processed requests, recording latencies as request passes through processing components inside app, exposing latencies of dependencies)
- Implement at least one of each metric types (counter, histogram, gauge)
- Configure Prometheus (through operator) to collect these metrics
- Display composite system + application metrics in dashboards (ready-made)
Registration is mandatory to attend this workshop session - reserve your seat now!