Telegraf Operator

Supported OS: linux | windows | macOS

Telegraf Operator is an application designed to create and manage individual Telegraf instances in Kubernetes clusters. Use it to collect, process, and aggregate metrics from applications in your Kubernetes cluster and send them to Coralogix.

Once installed, Telegraf Operator watches for pods deployed with a specific set of pod annotations. The advantage of using Telegraf Operator is that you only have to define the input plugin configuration for Telegraf when creating the pod annotations. Telegraf Operator then sets the configuration for the entire cluster, avoiding the need to configure a metrics destination when deploying applications.

This guide shows you how to run Telegraf Operator in Kubernetes to export your data to Coralogix.

Prerequisites

1. Sign up for a Coralogix account. Set up your account on the Coralogix domain corresponding to the region within which you would like your data stored.

2. Access your Coralogix private key.

3. Install Kubernetes. This should include installation of the command-line tool kubectl, designed to operate on your Kubernetes cluster.

4. Install and configure Helm. We suggest you use this guide to familiarize yourself with the basics of using Helm to manage packages on your Kubernetes cluster.

Set Up

Installation

STEP 1: Add the Influxdata Helm Chart Repository

Open a new terminal and install the Influxdata Helm chart repository:

helm repo add influxdata <https://helm.influxdata.com/>

helm repo update

STEP 2: Install Telegraf Operator

Update the Helm chart from the added repository:

export NAMESPACE=<namespace>

helm upgrade --install telegraf-operator influxdata/telegraf-operator -n $NAMESPACE

Configuration

STEP 1: Create YAML-Formatted Override File

Telegraf Operator allows you to configure how Telegraf sidecars send data using classes. Each class is a subset of the Telegraf configuration and defines where Telegraf Operator should be sending its outputs. You can create or modify existing classes using Helm values file, as shown in the example below.

In order to send your data to Coralogix, you are required to declare particular variables in your template:

• service_address: In order to send metrics to Coralogix, you will need to include your account’s specific domain in the Coralogix endpoint: otel-metrics.<domain>:443.
• private_key: Access your Coralogix private key. Your key is recorded in the override file as a secret in order to ensure that this sensitive information remains protected and unexposed.
• $HOSTNAME & $NODENAME : Each Telegraf Operator class includes these global tags. $HOSTNAME, which will serve as your Coralogix subsystem name, is typically set to pod name. $NODENAME is typically set to node name. Find out more about defining environment variables for Kubernetes containers here.
• $NAMESPACE: This variable is Telegraf Operator pod annotation telegraf.influxdata.com/env-fieldref-NAMESPACE: metadata.namespace that can be used to configure the Telegraf sidecar. This variable will serve as your Coralogix application name.

image:
 sidecarImage: "docker.io/library/telegraf:1.24.2"
classes:
  secretName: "telegraf-operator-classes"
  default: "infra"
  data:
    infra: |
      [[outputs.opentelemetry]]
        service_address = "<coralogix_endpoint>"
        compression = "gzip"
        [outputs.opentelemetry.coralogix]
          private_key = "<private_key>"
          application = "$NAMESPACE"
          subsystem = "$HOSTNAME"
      [global_tags]
        env = "ci"
        hostname = "$HOSTNAME"
        nodename = "$NODENAME"
        namespace = "$NAMESPACE"
        type = "infra"
        

Save this file as a coralogix-values.yaml file in a directory of your choice.

STEP 2: Apply the Configuration

Apply the configuration by running the following command:

helm upgrade --install telegraf-operator influxdata/telegraf-operator --values <path/to/coralogix-values.yaml> -n <namespace>

Validation

To test your configuration, deploy a Redis instance with Telegraf Operator annotations.

1. Create a pod configuration redis.yaml file with the following data:

apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: redis
spec:
  selector:
    matchLabels:
      app: redis
  serviceName: redis
  template:
    metadata:
      labels:
        app: redis
      annotations:
        telegraf.influxdata.com/inputs: |+
          [[inputs.redis]]
            servers = ["tcp://localhost:6379"]
        telegraf.influxdata.com/class: infra
        telegraf.influxdata.com/env-fieldref-NAMESPACE: metadata.namespace
    spec:
      containers:
      - name: redis
        image: redis:alpine3.16

2. Run the pod by executing:

kubectl apply -f <path/to/redis.yaml> -n <redis-namespace>

3. The pod should contain 2 running containers: redis and telegraf. Verify this by executing:

kubectl describe pod -n <redis-namespace> redis-0

Here is an output example:

4. Verify that the pod is running:

kubectl get sts -n <redis-namespace> redis

Here is an output example:

5. Access your Coralogix – Grafana dashboard to query the monitored targets.

• On the right-hand corner of your dashboard, click on the Grafana drop-down tab.

• Once you’ve accessed your Coralogix-Grafana dashboard, click on Explore tab in the left-hand browser.

• Click on the drop-down arrow of the Metrics browser and input target_info in the “Select a Metric” column.

• Click on the “Use Query” button in the bottom left-hand corner.

6. The query above should return metrics similar to the following:

target_info{__meta_applicationname="<redis-namespace>", __meta_subsystem="redis-0"}

7. To discover all Redis metrics, use the following PromQL:

{__meta_applicationname="<redis-namespace>", __meta_subsystem="redis-0"}

8. This should result in:

...
redis_total_connections_received{__meta_applicationname="<redis-namespace>, __meta_subsystem="redis-0", cx_tenant_id="32680", env="ci", host="redis-0", hostname="redis-0", namespace="<redis-namespace>", nodename="rock64-1", port="6379", replication_role="master", server="localhost", type="infra"}

Here is a display of the expected result:

9. Once you have tested and validated your configuration with the Redis instance, delete this service by running the following command:

kubectl delete -f <path/to/redis.yaml> -n <redis-namespace>

Additional References

Github: Official Telegraf-Operator Documentation

Github: Offical Influxdata Chart Repo

Support

Need help?

Our world-class customer success team is available 24/7 to walk you through your setup and answer any questions that may come up.

Feel free to reach out to us via our in-app chat or by sending us an email to [email protected].