Quickstart: Collector for application data

This tutorial demonstrates how to use the OpenTelemetry Collector to send application telemetry to Cloud Observability.

You will run a simple containerized application locally that will send trace data to a local containerized instance of the OpenTelemetry Collector, that will in turn send the trace data to Cloud Observability.

otel-collector-example

For more on the OpenTelemetry Collector, see the official OpenTelemetry docs.

If you’re deploying the collector in Kubernetes, we recommend using the Kubernetes Operator.

For sending metrics to Cloud Observability using the OpenTelemetry Collector, see Ingest metrics using the OpenTelemetry Collector.

Pre-Requisites

Docker and Docker Compose
A Cloud Observability account.
A Cloud Observability access token for the Cloud Observability project you would like to use.

Although instrumented code can send data directly to an Observability back-end (e.g. Cloud Observability) without a Collector, it is considered a best practice to send all OpenTelemetry data to your back-end via the OpenTelemetry Collector.

Tutorial: Running the OpenTelemetry Collector Locally

Clone the OpenTelemetry Examples repo.

 git clone https://github.com/lightstep/opentelemetry-examples.git

Edit the OpenTelemetry Collector Config file.

 cd opentelemetry-examples

Open collector/vanilla/collector.yaml for editing using your favourite editor.

It’s recommended that you make a copy of this file collector.yaml and save it as otelcol-lightstep.yml.

The file looks like this:

 receivers:
   otlp:
     protocols:
       grpc:
       http:

 exporters:
   logging:
     logLevel: debug
   otlp/ls:
     endpoint: ingest.lightstep.com:443
     headers: 
       "lightstep-access-token": "${LIGHTSTEP_ACCESS_TOKEN}"

 processors:
   batch:

 service:
   pipelines:
     traces:
       receivers: [otlp]
       processors: [batch]
       exporters: [logging, otlp/ls]

     metrics:
         receivers: [otlp]
         processors: [batch]
         exporters: [logging,otlp/ls]

Replace ${LIGHTSTEP_ACCESS_TOKEN} with your own Cloud Observability Access Token, and save the file. The access token tells what Cloud Observability project to send your telemetry data to.

A few noteworthy items:

The Collector can ingest data using both HTTP and gRPC. The receivers configuration may appear to be empty; however, it actually means that we are using the defult values for the receivers config. It is actually the equivalent of:

 receivers:
   otlp:
     protocols:
       grpc:
         endpoint: 0.0.0.0:4317
       http:
         endpoint: 0.0.0.0:4318

Cloud Observability ingests data in native OpenTelemetry Protocol (OTLP) format, the OTLP Exporter is used. The exporter can be called either otlp or follow the naming format otlp/<something>. In this example, the otlp/<something> format is added for clarity, to indicate that we are using OTLP to send data to Cloud Observability.
Though not mandatory, we are also using a Logging Exporter. This is helpful, as it prints our traces to the Collector’s stdout. In this case it’s your terminal.
We must define a pipeline in the service.pipelines section of the YAML config. Specifically, we need to define a pipeline for our traces. The pipeline tells the Collector:
- Where it’s getting trace data from (it’s being sent via OTLP)
- If there’s any processing that needs to be done (this is optional)
- Where to send data to. In our case, it’s to stdout (via the Logging Exporter) and to Cloud Observability (via OTLP Exporter)

If you find that you can’t see data in Cloud Observability, make sure that you’ve defined a pipeline, and that the pipeline’s exporter (for example service.pipelines.traces.exporters) lists your otlp/ls exporter.

Launch the Collector.

First, set up Docker networking. This is a one-time setup, and is done so that our sample app container can talk to the OpenTelemetry Collector container.

Open a new terminal window in the opentelemetry-examples folder, and run the following commands:

 docker network create --driver=bridge -o "com.docker.network.bridge.enable_icc"="true" otel-collector-demo

Now, in the same terminal window run the Collector. Note how it is using the newly-created otel-collector-demo network.

 docker run -it --rm \
     -p 4317:4317 \
     -p 4318:4318 \
     -v $(pwd)/collector/vanilla/otelcol-lightstep.yaml:/otel-config.yaml \
     --network="otel-collector-demo" \
     -h otel-collector \
     --name otel-collector \
     otel/opentelemetry-collector-contrib:0.53.0  \
     "/otelcol-contrib" \
     "--config=otel-config.yaml"

A few noteworthy items about the above command. It:

Exposes the OpenTelemetry Collector’s gRPC port (4317) and HTTP port (4318) for ingesting data in OTLP format. Code instrumented with OpenTelemetry will be in the OTLP format.
Maps the local otelcol-lightstep.yml file that you modified earlier, to the otel-config.yaml file in the container’s internal filesystem
Tells the OpenTelemetry Collector where to look for config YAML file via the --config flag.

Ensure that you are running the docker command from the opentelemetry-examples folder.

Launch the sample app.

Open a new terminal window and run the command below. Note how we are using the same network here as we did for the OpenTelemetry Collector, otel-collector-demo so that the app can reach the Collector.

 docker run -it --rm -h go-sample-server \
   --network="otel-collector-demo" \
   -p 9000:9000 ghcr.io/lightstep/go-sample-server:1.0.0

You should see output that looks something like this: otel-collector-sample-app-output

Call the service.

Open a new terminal window, and run the following:
```
1
 curl http://localhost:9000
```
You should see output that looks something like this:

When you return to the terminal window running the OpenTelemetry Collector, you’ll notice that the Logging Exporter has printed your trace to stdout:

When you return to the terminal windows running the sample app, you’ll notice the following new output:
See the Traces in Cloud Observability.

Log into Cloud Observability. You’ll be able to see the services listed in the Service Directory.

To view traces in your Cloud Observability project, click Explorer in the left navigation bar, and then click on any span in the Trace Analysis table.

For a more detailed trace exploration, check out Cloud Observability Notebooks.

Troubleshooting

The default OTLP Exporter from a Collector enables gzip compression and TLS. Depending on your network configuration, you may need to enable or disable certain other gRPC features. This page contains a complete list of configuration parameters for the Collector gRPC client.

In the event that you are unable to establish a gRPC connection to the Cloud Observability Observability platform, you can use the grpcurl tool to ensure connectivity from your network to our public satellites. Run the following command, replacing <YOUR_ACCESS_TOKEN> with your project’s access token:

grpcurl -H 'lightstep-access-token:<YOUR_ACCESS_TOKEN>' ingest.lightstep.com:443 list

You should see the following output, or something similar:

grpc.reflection.v1alpha.ServerReflection
jaeger.api_v2.CollectorService
lightstep.collector.CollectorService
lightstep.egress.CollectorService
opentelemetry.proto.collector.trace.v1.TraceService

If you do not see this output, or the request hangs, then something is blocking gRPC traffic from transiting your network to ours. Please ensure that any proxies are passing through the lightstep-access-token header.

For additional troubleshooting recommendations, see Troubleshooting Missing Data in Cloud Observability.

Quickstart: Collector for application data

Pre-Requisites

Tutorial: Running the OpenTelemetry Collector Locally

Troubleshooting

See also