Prometheus Agent

Send your data to Coralogix using Prometheus Agent, a new operational mode of running Prometheus, built directly into the Prometheus binary. The agent mode optimizes the remote write use case configuring the Prometheus instance while disabling some of Prometheus’ usual features – querying and alerting.

Use agent mode to:

• Focus on scraping metrics and sending them to Coralogix, with querying and alerting disabled
• Enable easier horizontal scalability for ingestion compared to server-mode
• Increase efficiency by replacing local storage with a customized TSDB WAL
• Improve scalability by enabling horizontal scalability for ingestion

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 and configure Prometheus Operator. Prometheus Agent collects ServiceMonitors and PodMonitors, which are enabled only when using the Prometheus Operator.

Note:

• The Prometheus operator must be in v0.59.0 at least in order to support the agent mode.

Configuration & Installation

STEP 1. Create a secret private key.

Create a Kubernetes secret for your Coralogix private key called coralogix-keys with the value PRIVATE_KEY.

Take this step in order to ensure that your private key remains protected and unexposed. While other methods exist for protecting this private information, we recommend creating a secret in this manner while running on a Kubernetes cluster.

Your private key, as well as the Helm chart, should be saved in the same namespace.

kubectl create secret generic coralogix-keys \\
  --from-literal=PRIVATE_KEY=<private-key>

The created secret should look like this:

apiVersion: v1
data:
  PRIVATE_KEY: <encrypted-private-key>
kind: Secret
metadata:
  name: coralogix-keys
  namespace: <the-release-namespace>
type: Opaque

STEP 2. Configure your Coralogix endpoints.

Using the region associated with your Coralogix domain, configure your Coralogix endpoint.

STEP 3. Add the Coralogix Helm charts repository.

Add the Coralogix Helm charts repository to the local repos list with the following command:

helm repo add coralogix-charts-virtual https://cgx.jfrog.io/artifactory/coralogix-charts-virtual

To get the updated Helm charts from the added repository, run:

helm repo update

STEP 4. Create an override.yaml file.

Create an override.yaml file, which includes the following:

---
# override.yaml:
prometheus:
  prometheusSpec:
    remoteWrite:
    - authorization:
        credentials:
          name: coralogix-keys
          key: PRIVATE_KEY
      name: prometheus-agent-coralogix
      queueConfig:
        capacity: 2500
        maxSamplesPerSend: 1000
        maxShards: 200
      remoteTimeout: 120s
      url: <YOUR_ENDPOINT>

STEP 5. Install the chart.

helm upgrade prometheus-agent coralogix-charts-virtual/prometheus-agent-coralogix \
  --install \
  --namespace=<your-namespace> \
  --create-namespace \
  -f override.yaml

Best Practices

By default, the Prometheus agent uses ephemeral volumes, which is not suitable for a production environment.

For a production environment, we highly recommended defining a persistent volume to avoid data loss between restarts. This is done by using the specification available on the values file.

Note:

If you do not specify the storageClassName, Kubernetes will use the default storage class available on the cluster.

prometheus:
  prometheusSpec:
    storageSpec:
      volumeClaimTemplate:
        spec:
          accessModes: ["ReadWriteOnce"]
          resources:
            requests:
              storage: 50Gi

Additional Resources

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 at [email protected].