Hot R.O.D. - Rides on Demand

This is a demo application that consists of several microservices and illustrates the use of the OpenTelemetry API & SDK. It can be run standalone, but requires Jaeger backend to view the traces. A tutorial / walkthrough is available:

• as a blog post Take Jaeger for a HotROD ride,
• as a video OpenShift Commons Briefing: Distributed Tracing with Jaeger & Prometheus on Kubernetes.

As of Jaeger v1.42.0 this application was upgraded to use the OpenTelemetry SDK for traces.

Features

• Discover architecture of the whole system via data-driven dependency diagram
• View request timeline & errors, understand how the app works
• Find sources of latency, lack of concurrency
• Highly contextualized logging
• Use baggage propagation to
  • Diagnose inter-request contention (queueing)
  • Attribute time spent in a service
• Use opentelemetry-go-contrib open source libraries to instrument HTTP and gRPC requests with minimal code changes

Running

Run everything via docker-compose

• Download docker-compose.yml from https://github.com/jaegertracing/jaeger/blob/main/examples/hotrod/docker-compose.yml
• Optional: find the latest Jaeger version (see https://www.jaegertracing.io/download/) and pass it via environment variable JAEGER_VERSION. Otherwise docker compose will use the latest tag, which is fine for the first time you download the images, but once they are in your local registry the latest tag is never updated and you may be running stale (and possibly incompatible) verions of Jaeger and the HotROD app.
• Run Jaeger backend and HotROD demo, e.g.:
  • JAEGER_VERSION=1.52 docker-compose -f path-to-yml-file up
• Access Jaeger UI at http://localhost:16686 and HotROD app at http://localhost:8080
• Shutdown / cleanup with docker-compose -f path-to-yml-file down

Alternatively, you can run each component separately as described below.

Run everything in Kubernetes

kustomize build ./kubernetes | kubectl apply -f -
kubectl port-forward -n example-hotrod service/example-hotrod 8080:frontend
# In another terminal
kubectl port-forward -n example-hotrod service/jaeger 16686:frontend

# To cleanup
kustomize build ./kubernetes | kubectl delete -f -

Access Jaeger UI at http://localhost:16686 and HotROD app at http://localhost:8080

Run Jaeger backend

An all-in-one Jaeger backend is packaged as a Docker container with in-memory storage.

docker run \
  --rm \
  --name jaeger \
  -p6831:6831/udp \
  -p16686:16686 \
  -p14268:14268 \
  jaegertracing/all-in-one:latest

Jaeger UI can be accessed at http://localhost:16686.

Run HotROD from source

git clone git@github.com:jaegertracing/jaeger.git jaeger
cd jaeger
go run ./examples/hotrod/main.go all

Run HotROD from docker

docker run \
  --rm \
  --link jaeger \
  --env OTEL_EXPORTER_JAEGER_ENDPOINT=http://jaeger:14268/api/traces \
  -p8080-8083:8080-8083 \
  jaegertracing/example-hotrod:latest \
  all

Then open http://127.0.0.1:8080

Metrics

The app exposes metrics in either Go's expvar format (by default) or in Prometheus format (enabled via -m prometheus flag).

• expvar: curl http://127.0.0.1:8083/debug/vars
• Prometheus: curl http://127.0.0.1:8083/metrics

Linking to traces

The HotROD UI can generate links to the Jaeger UI to find traces corresponding to each executed request. By default it uses the standard Jaeger UI address http://localhost:16686, but if your Jaeger UI is running at a different address, it can be customized via -j <address> flag passed to HotROD, e.g.

go run ./examples/hotrod/main.go all -j http://jaeger-ui:16686