Send AKS metrics with Helm and the OpenTelemetry Collector

• Overview
• Standard configuration Linux
• Standard configuration Windows
• Customizing Helm chart parameters
• Uninstalling the Chart

Overview

logzio-otel-k8s-metrics allows you to ship metrics from your Kubernetes cluster to Layerlog with the OpenTelemetry collector. For AKS clusters, this chart also allows you to ship Windows node metrics.

This chart is a fork of the opentelemetry-collector Helm chart. The main repository for Layerlog helm charts are logzio-helm.

It is also dependent on the kube-state-metrics, prometheus-node-exporter and prometheus-pushgateway charts, which are installed by default.

To disable the dependency during installation, set kubeStateMetrics.enabled, nodeExporter.enabled or pushGateway.enabled to false.

Sending logs from nodes with taints

If you want to ship logs from any of the nodes that have a taint, make sure that the taint key values are listed in your in your daemonset/deployment configuration as follows:

tolerations:
- key: 
  operator: 
  value: 
  effect: 

To determine if a node uses taints as well as to display the taint keys, run:

kubectl get nodes -o json | jq ".items[]|{name:.metadata.name, taints:.spec.taints}"

Standard configuration for Linux nodes

Add logzio-helm repo to your helm repo list

  helm repo add logzio-helm https://logzio.github.io/logzio-helm
  helm repo update

Deploy the Helm chart

1. Configure the relevant parameters in the following code:

   helm install --namespace <<YOUR-NAMESPACE>>  \
   --set secrets.MetricsToken=<<METRICS-SHIPPING-TOKEN>> \
   --set secrets.ListenerHost="https://<<LISTENER-HOST>>:8053" \
   --set secrets.p8s_logzio_name=<<ENV-TAG>> \
   logzio-otel-k8s-metrics logzio-helm/logzio-otel-k8s-metrics
   

   • Replace <<YOUR_NAMESPACE>> with the required namespace.

   • Replace <<METRICS-SHIPPING-TOKEN>> with a token for the Metrics account you want to ship to.
     Look up your Metrics token.

   • Replace <<LISTENER-HOST>> with the host for your region. For example, listener.layerlog.com if your account is hosted on AWS US East, or listener-nl.layerlog.com if hosted on Azure West Europe.

   • Replace <<ENV-TAG>> with the name for the environment’s metrics, to easily identify the metrics for each environment.

2. Run the code.

Check Layerlog for your metrics

Give your metrics some time to get from your system to ours.

Log in to your Layerlog account and navigate to the current instructions page inside the Layerlog app. Install the pre-built dashboard to enhance the observability of your metrics.

To view the metrics on the main dashboard, log in to your Layerlog Metrics account, and open the Layerlog Metrics tab.

Standard configuration for Windows nodes

Add logzio-helm repo to your helm repo list

  helm repo add logzio-helm https://logzio.github.io/logzio-helm
  helm repo update

Update your Windows node pool credentials (if needed)

To extract and scrape metrics from Windows nodes, you need to install a Windows Exporter service on the node host. To do this, you need to establish a SSH connection to the node by authenticating with a username and password. The windows-exporter-installer job performs the installation on each Windows node using the provided Windows credentials. The default username for Windows node pool is azureuser.

If your Windows node pool does not share the same username and password across the nodes, you will need to run the windows-exporter-installer job for each node pool using the relevant credentials. You can change your Windows node pool password in AKS cluster with the following command:

   az aks update \
   --resource-group $RESOURCE_GROUP \
   --name $CLUSTER_NAME \
   --windows-admin-password $NEW_PW

• Replace RESOURCE_GROUP with the resource group name.
• Replace CLUSTER_NAME with the cluster name.
• Replace NEW_PW with the password selected for this pool.

Deploy the Helm chart

1. Configure the relevant parameters in the following code:

   helm install --namespace <<YOUR-NAMESPACE>>  \
   --set secrets.MetricsToken=<<METRICS-SHIPPING-TOKEN>> \
   --set secrets.ListenerHost="https://<<LISTENER-HOST>>:8053" \
   --set secrets.p8s_logzio_name=<<ENV-TAG>> \
   --set secrets.windowsNodeUsername=<<WINDOWS-NODE-USERNAME>> \
   --set secrets.windowsNodePassword=<<WINDOWS-NODE-PASSWORD>> \
   logzio-otel-k8s-metrics logzio-helm/logzio-otel-k8s-metrics
   

   • Replace <<YOUR_NAMESPACE>> with the required namespace.

   • Replace <<METRICS-SHIPPING-TOKEN>> with a token for the Metrics account you want to ship to.
     Look up your Metrics token.

   • Replace <<LISTENER-HOST>> with the host for your region. For example, listener.layerlog.com if your account is hosted on AWS US East, or listener-nl.layerlog.com if hosted on Azure West Europe.

   • Replace <<ENV-TAG>> with the name for the environment’s metrics, to easily identify the metrics for each environment.

   • Replace <WINDOWS-NODE-USERNAME>> with the username for the Node pool you want the Windows Exporter to be installed on.

   • Replace <<WINDOWS-NODE-PASSWORD>> with the password for the Node pool you want the Windows exporter to be installed on.

2. Run the code.

Check Layerlog for your metrics

Give your metrics some time to get from your system to ours.

Log in to your Layerlog account and navigate to the current instructions page inside the Layerlog app. Install the pre-built dashboard to enhance the observability of your metrics.

To view the metrics on the main dashboard, log in to your Layerlog Metrics account, and open the Layerlog Metrics tab.

Customizing Helm chart parameters

Configure customization options

You can use the following options to update the Helm chart parameters:

• Specify parameters using the --set key=value[,key=value] argument to helm install

• Edit the values.yaml

• Overide default values with your own my_values.yaml and apply it in the helm install command.

Example:

helm install logzio-otel-k8s-metrics logzio-helm/logzio-otel-k8s-metrics -f my_values.yaml 

Customize the metrics collected by the Helm chart

The default configuration uses the Prometheus receiver with the following scrape jobs:

• Cadvisor: Scrapes container metrics
• Kubernetes service endpoints: These jobs scrape metrics from the node exporters, from Kube state metrics, from any other service for which the prometheus.io/scrape: true annotaion is set, and from services that expose Prometheus metrics at the /metrics endpoint.

To customize your configuration, edit the config section in the values.yaml file.

Uninstalling the Chart

The uninstall command is used to remove all the Kubernetes components associated with the chart and to delete the release.

To uninstall the logzio-otel-k8s-metrics deployment, use the following command:

helm uninstall logzio-otel-k8s-metrics