Traceableai/traceableai_traefik_plugin

Traceable Traefik Plugin

Configuration

To configure the Traceable Traefik plugin you will need a Traceable Platform Agent deployed and configured. In Kubernetes, you could deploy TME as a sidecar to your traefik deployment.

In your traefik.yml you need to add the plugin configuration:

experimental:
  plugins:
    plugin-traceableai:
      moduleName: github.com/Traceableai/traceable_traefik_plugin
      moduleVersion: 1.0.0

Then add the plugin as a middleware & configure it on any router you want to collect data for:

http:
  routers:
    whoami-router:
      rule: "Host(`echo-service.docker.localhost`)"
      service: echo-service
      middlewares:
        - my-plugin-traceableai
  
  
  middlewares:
    my-plugin-traceableai:
      plugin:
        plugin-traceableai:
          allowedContentTypes:
            - "json"
            - "xml"
            - "x-www-form-urlencoded"
            - "grpc"
          bodyCaptureSize: 131072
          serviceName: traefik
          tpaEndpoint: http://REPLACE_ME_WITH_TPA_HOST_OR_IP:5442

Local Testing

To test traefik locally, first navigate to the local directory.

Run docker-compose up reverse-proxy to start traefik.

In another terminal, run docker-compose up echo-service to start a echo server.

You can access the traefik dashboard at http://localhost:8080.

You can access the echo service by sendina Host header set to echo-service.docker.localhost. Example: curl -H "Host: echo-service.docker.localhost" http://localhost:80

To test grpc, run docker-compose up grpc-server to start a grpc server.

Then use the example grpcurl_example.sh, make sure you run the script from the local/grpc directory.

ex: cd ./local/grpc && ./grpcurl_example.sh