docker-otel-lgtm

An OpenTelemetry backend in a Docker image.

The grafana/otel-lgtm Docker image is an open source backend for OpenTelemetry that’s intended for development, demo, and testing environments. If you are looking for a production-ready, out-of-the box solution to monitor applications and minimize MTTR (mean time to resolution) with OpenTelemetry and Prometheus, you should try Grafana Cloud Application Observability.

Documentation

• Blog post: An OpenTelemetry backend in a Docker image: Introducing grafana/otel-lgtm

Get the Docker image

The Docker image is available on Docker hub: https://hub.docker.com/r/grafana/otel-lgtm

Run the Docker image

# Unix/Linux
./run-lgtm.sh

# Windows (PowerShell)
./run-lgtm

Configuration

Enable logging

You can enable logging for troubleshooting:

Environment Variable	Enable Logging in
ENABLE_LOGS_GRAFANA	Grafana
ENABLE_LOGS_LOKI	Loki
ENABLE_LOGS_PROMETHEUS	Prometheus
ENABLE_LOGS_TEMPO	Tempo
ENABLE_LOGS_OTELCOL	OpenTelemetry Collector
ENABLE_LOGS_ALL	all of the above

This has nothing to do with the application logs, which are collected by OpenTelemetry.

Run lgtm in kubernetes

# create k8s resources
kubectl apply -f k8s/lgtm.yaml

# port forwarding
kubectl port-forward service/lgtm 3000:3000 4317:4317 4318:4318

Send OpenTelemetry Data

There's no need to configure anything: The Docker image works with OpenTelemetry's defaults.

# Not needed as these are the defaults in OpenTelemetry:
export OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318

View Grafana

Log in to http://localhost:3000 with user admin and password admin.

Build the Docker image from scratch

cd docker/
docker build . -t grafana/otel-lgtm

Build and run the example app

Run the example REST service:

# Unix/Linux
./run-example.sh

# Windows (PowerShell)
./run-example

Generate traffic:

# Unix/Linux
./generate-traffic.sh

# Windows (PowerShell)
./generate-traffic

Run example apps in different languages

The example apps are in the examples/ directory. Each example has a run.sh or run.cmd script to start the app.

Every example implements a rolldice service, which returns a random number between 1 and 6.

Each example uses a different application port (to be able to run all applications at the same time).

Example	Service URL
Java	curl http://localhost:8080/rolldice
Go	curl http://localhost:8081/rolldice
Python	curl http://localhost:8082/rolldice
dotnet	curl http://localhost:8083/rolldice

Related Work

• Metrics, Logs, Traces and Profiles in Grafana: https://github.com/grafana/intro-to-mltp