Docker based demo application for generating Prometheus exemplar data that generates traces in an Azure Application Insights workspace.
Docker Compose is required to run the containers in this example.
The following environment variables are also required:
APPLICATIONINSIGHTS_CONNECTION_STRING- The connection string for the Application Insights workspace that traces will be written to.TENANT_ID- The tenant ID for an App Registration that will be used to authenticate to Azure.CLIENT_ID- The client ID for an App Registration that will be used to authenticate to Azure.CLIENT_SECRET- The client secret for an App Registration that will be used to authenticate to Azure.
Export the above environment variables and then run ./start.sh from a terminal. The docker containers will be created and Grafana will be available on localhost:3002.
The Grafana instance will be preconfigured with Azure Monitor, Prometheus, Loki, and Tempo data sources.
These data sources are appropriately configured to connect to the relevant container instances where the generated data can be viewed.
Querying the prometheus_exemplars_bucket metric in the Prometheus data source with exemplars enabled will display metrics that should have exemplars attached to them. Please note that there may be a delay between a metric being populated and the exemplar becoming available in Azure.
To stop the demo application run ./stop.sh.