In this talk, I give an overview of what OpenTelemetry (OTeL) is and what it can help you to achieve. The example consists of multiple microservices which are deployed in a Kubernetes cluster fully traced with OTeL. The traced components are:
- React frontend
- Kubernetes Ingress Controller
- API backend
- Kafka
- Asynchronous processing microservice
Repository: https://github.com/secustor/opentelemetry-meetup
Components
Snapshot
Snapshot is a React photo gallery to showcase capabilities. In this example, OpenTelemetry tracing capabilities have been added to allow an easy way to demonstrate the capture of user input and API call tracing.
Code: https://github.com/secustor/opentelemetry-meetup/tree/master/apps/snapshot
Report
The report consists of the Kafka producer
which accepts API calls from Snapshot and writes the request content on a Kafka topic.
From this Kafka topic, the consumer reads the received messages asynchronously and prints them.
Code: https://github.com/secustor/opentelemetry-meetup/tree/master/apps/report
OpenTelemetry
The example makes use of the OpenTelemetry Collector and SDKs for collecting, transforming as well as loading and for instrumentation respectively.
Observability Stack
The Stack below is used to store Observability signals ( Traces, Metrics, Logs ) and graph them:
- Grafana ( Dashboarding )
- Tempo ( Traces )
- Loki ( Logs)
- Prometheus ( Metrics )