This guide explains how to install the public StackBooster Agent chart in your Kubernetes cluster using Helm.

Important notes:

• You will install the public chart from the StackBooster Helm repository.
• The agent requires a default StorageClass to be present in the cluster.
• The agent requires cert-manager to be installed (CRDs must be enabled).

Prerequisites #

• EKS Kubernetes cluster (v1.32+ recommended)
• kubectl and Helm v3 installed
• Cluster admin permissions to install CRDs and annotate StorageClass

Add Helm Repositories #

helm repo add jetstack https://charts.jetstack.io
helm repo add stackbooster https://raw.githubusercontent.com/stackbooster/helm/main/charts
helm repo update

Install cert-manager #

Install cert-manager and its CRDs (required by StackBooster):

echo "Install cert manager"
helm install cert-manager jetstack/cert-manager \
  --namespace cert-manager \
  --create-namespace \
  --version v1.17.0 \
  --set crds.enabled=true

Verify cert-manager:

kubectl get pods -n cert-manager

Ensure a Default StorageClass Exists #

StackBooster requires a default StorageClass. If your cluster already has a default StorageClass, you can skip this step:

kubectl get storageclass

If you need to set an existing StorageClass (for example, gp2 on AWS) as default:

kubectl annotate storageclass gp2 storageclass.kubernetes.io/is-default-class=true --overwrite=true

Replace gp2 with your StorageClass name if different (e.g., gp3, standard, etc.).

Install StackBooster Agent #

Install the public StackBooster Agent chart (example uses version 1.0.10):

echo "Install stackbooster"
helm upgrade --install stackbooster stackbooster/stackbooster-agent \
  --namespace stackbooster \
  --create-namespace \
  --version 1.0.10

How to Customize Values #

Use –set flags for a small number of overrides, for example:

helm upgrade --install stackbooster stackbooster/stackbooster-agent \
  --namespace stackbooster \
  --create-namespace \
  --version 1.0.10 \
  --set agentMode=readonly \
  --set cloudProvider=aws

For metrics endpoints configuration (Prometheus, VictoriaMetrics) see metricsConfig-guide.md in this directory, and replicate the settings using either sb-values.yaml or --set flags.

Verify Installation #

kubectl get pods -n stackbooster
kubectl get svc -n stackbooster
kubectl get crds | grep stackbooster || true

You should see StackBooster controller/agent components running without CrashLoopBackOff.

Upgrades #

To upgrade to a newer chart version later:

helm repo update
helm upgrade stackbooster stackbooster/stackbooster-agent \
  --namespace stackbooster \
  --version <new-version> \
  -f sb-values.yaml   # if you used a values file

Uninstall #

helm uninstall stackbooster -n stackbooster

About OpenTelemetry and Instrumentation Layer #

• In chart version 1.0.10, customers install the agent as shown above.
• Beginning with chart version 1.0.11 (when published), OTEL support is included in the chart, but the application instrumentation layer must be deployed separately by customers.
• For guidance on deploying the instrumentation layer (when you move to 1.0.11+), see the OTEL guide: OpenTelemetry Deployment and Integration Guide.

Troubleshooting #

• cert-manager pods not ready: ensure --set crds.enabled=true was used and all pods in cert-manager namespace are Running.
• No default StorageClass: annotate your preferred StorageClass as default (see step above) or create one if none exists.
• Pods CrashLooping: check logs with kubectl logs <pod> -n stackbooster and review your overrides in sb-values.yaml.