Quickstart Kubernetes: Collector and Operator

This tutorial demonstrates how to use the Kubernetes Operator for OpenTelemetry Collector to send infrastructure metrics with Prometheus, and optionally application traces and logs, to Cloud Observability using a Helm chart following Collector general guidelines. Additionally, you’ll enable the OpAMP bridge, which is an OpAMP agent that sends configuration and status reports for the Collectors you are running, allowing you to view collector health in Cloud Observability.

Use this method only if you want an exact duplicate of your Prometheus installation, otherwise follow the general-purpose Kubernetes quickstart here.

A prerequisite of this quickstart is a running Kubernetes cluster. It can be either a standard Kubernetes distribution or a managed Kubernetes distribution like Azure AKS, Google GKE, or AWS EKS. If you’d just like to test locally, we recommend using minikube.

For more on the Kubernetes Operator for OpenTelemetry Collector, see the official OpenTelemetry docs.

Prerequisites

  • A Kubernetes cluster (either local using a tool like minikube or a cluster running in the cloud) with at least 2 CPUs and 4 GB of memory.
  • Helm v3 or later
  • A Cloud Observability account
  • A Cloud Observability access token for the Cloud Observability project you would like to use.
  • A Cloud Observability API key created specifically for the OpAMP agent.

    The OpAMP agent allows you to monitor Collector health in Cloud Observability.
    Set the scope to project, and assign it the OpAMP Agent role.

Verify your setup

  1. Run the following command to verify you are connected to a Kubernetes cluster.

    1
    
     kubectl cluster-info
    

    If you see errors or cannot connect, follow the instructions from minikube or your cloud provider on authenticating with your cluster.

  2. Next, verify Helm is installed.

    1
    
     helm version
    

    Verify you are on Helm v3.

We recommend using Helm to manage dependencies and upgrades. However, if you cannot deploy Helm charts, you can use the helm template command to automatically generate Kubernetes manifests from an existing chart.

Add Helm repositories and install charts

  1. Run the following command to add the following Helm respositories and pull the latest charts:

    These Helm charts configure your environment (especially the Collector) to work best with Cloud Observability.

    1
    2
    3
    4
    5
    
     helm repo add jetstack https://charts.jetstack.io
     helm repo add open-telemetry https://open-telemetry.github.io/opentelemetry-helm-charts
     helm repo add prometheus https://prometheus-community.github.io/helm-charts
     helm repo add lightstep https://lightstep.github.io/otel-collector-charts
     helm repo update
    
  2. Next, install the cert-manager charts on your cluster. The Cert Manager manages certificates needed by the Operator to subscribe to in-cluster Kubernetes events.

    1
    2
    3
    4
    5
    6
    
     helm install \
         cert-manager jetstack/cert-manager \
         --namespace cert-manager \
         --create-namespace \
         --version v1.8.0 \
         --set installCRDs=true
    
  3. Create a namespace for ServiceNow Cloud Observability

    1
    
     kubectl create ns sn-cloud-obs
    
  4. Install the OpenTelemetry Operator chart. The Operator automates the creation and management of collectors, autoscaling, code instrumentation, scraping metrics endpoints, and more.
    1
    2
    3
    4
    5
    
     helm install \
         opentelemetry-operator open-telemetry/opentelemetry-operator \
         -n sn-cloud-obs \
         --set "manager.collectorImage.repository=otel/opentelemetry-collector-k8s" \
         --version 0.56.0
    
  5. Run the following command to verify both charts successfully deployed with a status that says deployed:
    1
    
     helm list -A
    

Send Kubernetes metrics to Cloud Observability

Kubernetes has built-in support for hundreds of useful metrics that help teams understand the health of their containers, pods, nodes, workloads, and internal system components. Cloud Observability provides a Helm chart to automatically configure collectors to send these metrics to Cloud Observability.

  1. Create a secret that holds your Cloud Observability Access Token.

    1
    2
    
     export LS_TOKEN='<your-token>'
     kubectl create secret generic otel-collector-secret -n sn-cloud-obs --from-literal="LS_TOKEN=$LS_TOKEN"
    
  2. Create another secret that holds your Cloud Observability API key.

    1
    2
    
     export LS_OPAMP_API_KEY='<your-api-key>'
     kubectl create secret generic otel-opamp-bridge-secret -n sn-cloud-obs --from-literal="LS_OPAMP_API_KEY=$LS_OPAMP_API_KEY"
    
  3. Install the collector-k8s chart. This chart automatically creates collectors to pull Kubernetes metrics and send them to your Cloud Observability project. We recommend you also specify the name of your cluster when installing the chart, which your can use by setting the clusterName variable:

    1
    2
    3
    4
    5
    
     helm install kube-otel-stack lightstep/kube-otel-stack \
         -n sn-cloud-obs \
         --set metricsCollector.clusterName=your-cluster-name \
         # --set otlpDestinationOverride="ingest.eu.lightstep.com:443" \ # EU data center
         # --set opAMPBridge.endpoint="wss://opamp.eu.lightstep.com/v1/opamp" # EU data center  
    
  4. Verify the pods from the charts have been deployed with no errors:

    1
    
     kubectl get pods
    

    You should see pods for a node exporter, the operator, kube-state-metrics, and multiple collectors.

See metrics in Cloud Observability

In Cloud Observability, you can view your metrics in either a notebook or dashboard.

When using notebooks you can click on any Kubernets metrics in the all telemetry dropdown. Check the scrape_series_added metric first, which lets you know many Kubernetes metrics are being ingested.

For dashboards, there are several pre-built dashboards that display Kubernetes metrics. For example, to see Pod metrics, in the Dashboard view, click Create a pre-built dashboard, and choose “Kubernetes Comprehensive (Prometheus)”.

Send traces to Cloud Observability

Send data from your applications to Cloud Observability

You can also use the Operator to deploy a collector configured to send trace data to Cloud Observability. The chart configures a collector for tracing following general guidelines.

  1. Run the following command to deploy a new Collector configured for trace data into the cluster. Replace your-cluster-name with the name of the cluster you are connected to.

    1
    2
    3
    4
    5
    
     helm upgrade kube-otel-stack lightstep/kube-otel-stack \
       -n sn-cloud-obs --set tracesCollector.enabled=true \
       --set tracesCollector.clusterName=your-cluster-name
       # --set otlpDestinationOverride="ingest.eu.lightstep.com:443" \ # EU data center
       # --set opAMPBridge.endpoint=wss://opamp.eu.lightstep.com/v1/opamp # EU data center
    
  2. Next, verify that the Collector configured for tracing has been deployed:

    1
    
     kubectl get services
    

    You should see a new service with the name kube-otel-stack-traces-collector with ports 4317/TCP and 8888/TCP open.

  3. Configure your OpenTelemetry-instrumented applications running in the cluster to export traces to an OTLP/gRPC endpoint kube-otel-stack-traces-collector:4317. More information is available on how to instrument applications in the Quickstart: Instrumentation documentation or follow the instructions below to deploy the demo application.

The Operator, for languages like Java, .NET, Node, and Python, supports auto-instrumenting code running in clusters. This lets you deploy SDKs automatically without any code changes. More details are available in the OpenTelemetry Community Docs.

Send data from the OpenTelemetry Demo

If you don’t have existing services that are instrumented, you can deploy a demo microservice environment to your cluster maintained by the OpenTelemetry Community that uses the collectors and configuration you deployed using the kube-otel-stack Helm chart.

Before proceeding, we recommend creating a separate sandbox or development project for testing with non-production data. If you create a new project, you will need to update the access token value you set in previous steps.

  1. Create a new values.yaml with the following content. This configures the OpenTelemetry Demo Helm chart to send metrics and traces to the collectors deployed by the kube-otel-stack chart:

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    
     opentelemetry-collector:
       config:
         exporters:
           otlp/traces:
             endpoint: kube-otel-stack-traces-collector:4317
             tls:
               insecure: true
           otlp/metrics:
             endpoint: kube-otel-stack-metrics-collector:4317
             tls:
               insecure: true
         service:
           pipelines:
             metrics:
               receivers: [otlp]
               processors: [batch]
               exporters: [logging, otlp/metrics]
             traces:
               receivers: [otlp]
               processors: [batch]
               exporters: [logging, otlp/traces]
    
  2. Deploy the demo environment with your values.yaml file.

    1
    
     helm upgrade my-otel-demo open-telemetry/opentelemetry-demo --install -f values.yaml
    
  3. After a few minutes, you should see new services, spans, and metrics in your Cloud Observability project.

Send logs to Cloud Observability

Send Kubernetes Events and pod logs to Cloud Observability

You can also use the Operator to deploy a collector configured to send log data to Cloud Observability. The chart configures a collector for logging following general guidelines and will forward Kubernetes events and pod logs by default.

  1. Run the following command to deploy a new Collector configured for logging data into the cluster.

    1
    2
    3
    4
    5
    6
    
     helm upgrade kube-otel-stack lightstep/kube-otel-stack \
       -n sn-cloud-obs \
       --set tracesCollector.enabled=true \
       --set logsCollector.enabled=true
       # --set otlpDestinationOverride="ingest.eu.lightstep.com:443" \ # EU data center
       # --set opAMPBridge.endpoint="wss://opamp.eu.lightstep.com/v1/opamp" # EU data center
    
  2. Next, verify that the Collector configured for logging has been deployed:

    1
    
     kubectl get services
    

    You should see a new service with the name kube-otel-stack-logs-collector.

  3. After a few minutes, you should see logs in Cloud Observability.

Troubleshooting

The first thing you should do when troubleshooting collector issues is make sure data from your network can reach Cloud Observability. Your firewall or cloud configuration may be preventing a connection.

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 connection to the Cloud Observability platform, you can use curl to verify HTTP/2 connectivity to our collectors. Run the following command, replacing <YOUR_ACCESS_TOKEN> with your project’s access token:

1
2
curl -D- -XPOST --http2-prior-knowledge -H "lightstep-access-token: <YOUR_ACCESS_TOKEN>" https://ingest.lightstep.com/access-test # US data center
# curl -D- -XPOST --http2-prior-knowledge -H "lightstep-access-token: <YOUR_ACCESS_TOKEN>" https://ingest.eu.lightstep.com/access-test # EU data center

You should see the following output, or something similar:

1
2
3
4
5
6
7
HTTP/2 200
content-length: 2
content-type: text/plain
date: Thu, 09 May 2024 15:39:14 GMT
server: envoy

OK

If you do not see this output, or the request hangs, then something is blocking HTTP2 traffic from transiting your network to ours.

If you see HTTP/2 401, your request succeeded, but your token was not accepted. Some things to check:

  • Validity of your access token.
  • Ensure proxies are passing through the lightstep-access-token header.

Alternatively, to exercise the full gRPC request/response cycle, you can try emitting a single span to your project using the otel-cli tool. Refer to this example image and commands for running the CLI tool in Kubernetes and Docker on GitHub. Only send test spans to a non-production project.

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

See also

Use the OpenTelemetry Collector

OpenTelemetry Quickstart

Updated Apr 3, 2024