Helm Installation (K8s Installation)

What you'll learn: Deploy the Scoutflo Kubernetes Agent to enable AI-driven incident investigation and root-cause analysis across your Kubernetes infrastructure.

Overview

The Scoutflo Kubernetes Agent is a lightweight monitoring solution that provides deep visibility into your Kubernetes clusters. Once deployed, it automatically discovers and monitors your infrastructure to power intelligent incident response.

Prerequisites Checklist

Before starting the installation, ensure you have the required access and components:

✅ System Requirements

• Kubernetes 1.21+ (EKS, GKE, AKS, or self-managed)

• Helm 3.8+ installed and configured

• Active kubeconfig context for target cluster

• Ingress controller deployed (NGINX, Traefik, etc.)

✅ Access Requirements

Cluster Admin (Recommended)

You have cluster-admin privileges or equivalent permissions to:

• Create namespaces, deployments, services, ingresses

• Manage RBAC resources (ServiceAccounts, ClusterRoles, ClusterRoleBindings)

• Create and manage secrets

Limited Permissions

✅ DNS & TLS Requirements

• DNS Control: Ability to create/update DNS records for agent hostname

• TLS Management: Either cert-manager setup OR manual certificate management

• Domain Ownership: Access to domain where agent will be hosted

Important: If you don't control DNS or TLS certificates, coordinate with your platform/infra team before proceeding.

---

Installation Guide

1

Prepare your environment

Install Required Components

Ingress Controller

If you don't have an ingress controller, install NGINX:

# Add the official NGINX ingress repository
helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
helm repo update

# Install NGINX ingress controller
helm install ingress-nginx ingress-nginx/ingress-nginx \
  --namespace ingress-nginx \
  --create-namespace \
  --wait

Cert-Manager

Configure TLS Certificate Issuer

Create a ClusterIssuer for Let's Encrypt certificates:

# File: clusterissuer.yaml
apiVersion: cert-manager.io/v1
kind: ClusterIssuer
metadata:
  name: letsencrypt-prod
spec:
  acme:
    server: https://acme-v02.api.letsencrypt.org/directory
    email: [e-mail]  # Replace with your email
    privateKeySecretRef:
      name: letsencrypt-prod
    solvers:
      - http01:
          ingress:
            class: nginx  # Match your ingress class

# Apply the ClusterIssuer
kubectl apply -f clusterissuer.yaml

2

Configure Helm registry access

Registry Credentials: Scoutflo will provide you with private registry credentials. Contact your Scoutflo representative if you haven't received them.

# Login to Scoutflo's private Helm registry
helm registry login ghcr.io \
  --username <YOUR_SCOUTFLO_USERNAME> \
  --password <YOUR_SCOUTFLO_TOKEN>

3

Deploy the Scoutflo Agent

Ready to install! The following command will deploy the agent with all necessary components.

helm install scoutflo-k8s-agent \
  oci://ghcr.io/scoutflo/k8s-dashboard/k8s-dashboard \
  --version 0.1.277 \
  --namespace scoutflo-k8s-agent \
  --create-namespace \
  --set namespace.name=scoutflo-k8s-agent \
  --set ingress.hostname=api-k8sagent.your-domain.com \
  --set ingress.tls.wildcardHostname=your-domain.com \
  --set ingress.tls.secretName=letsencrypt-prod \
  --set image.dockerConfigSecret=<REGISTRY_SECRET_NAME> \
  --set auth.username=<SCOUTFLO_AGENT_USERNAME> \
  --set auth.password=<SCOUTFLO_AGENT_PASSWORD> \
  --set auth.sessionSecret=<RANDOM_SESSION_SECRET> \
  --set auth.deploymentId=<SCOUTFLO_DEPLOYMENT_ID> \
  --set auth.agentId=<SCOUTFLO_AGENT_ID> \
  --set rbac.agentImpersonation.userResourceNames=<SCOUTFLO_AGENT_USERNAME> \
  --wait

Parameter Reference

Parameter	Description	Example Value
ingress.hostname	FQDN where agent will be accessible	api-k8sagent.company.com
ingress.tls.wildcardHostname	Root domain for TLS certificate	company.com
auth.username	Agent authentication username	Provided by Scoutflo
auth.password	Agent authentication password	Provided by Scoutflo
auth.sessionSecret	Random session encryption key	Generate with openssl rand -hex 32
auth.deploymentId	Your Scoutflo deployment ID	Provided by Scoutflo
auth.agentId	Unique agent identifier	Provided by Scoutflo

Security Note: Replace all placeholder values (<...>) with actual values provided by the Scoutflo team. Never commit these values to version control.

4

Verify installation

Check that all components are running correctly:

# Check agent pods
kubectl get pods -n scoutflo-k8s-agent

# Verify ingress configuration
kubectl get ingress -n scoutflo-k8s-agent

# Check TLS certificate status
kubectl describe certificate -n scoutflo-k8s-agent

# View agent logs
kubectl logs -n scoutflo-k8s-agent deployment/scoutflo-k8s-agent --tail=50

Expected Output

✅ Healthy Installation:

NAME                                 READY   STATUS    RESTARTS   AGE
scoutflo-k8s-agent-xxxxxxxxx-xxxxx  1/1     Running   0          2m

NAME                     CLASS   HOSTS                           ADDRESS          PORTS     AGE
scoutflo-k8s-agent      nginx   api-k8sagent.your-domain.com    203.0.113.10     80, 443   2m

NAME                     READY   SECRET                  AGE
scoutflo-k8s-agent-tls  True    letsencrypt-prod        2m

5

Configure DNS

Create a DNS record pointing to your ingress controller's external IP:

# Get ingress external IP
kubectl get ingress scoutflo-k8s-agent -n scoutflo-k8s-agent \
  -o jsonpath='{.status.loadBalancer.ingress[0].ip}'

Create an A record:

• Name: api-k8sagent (or your chosen subdomain)

• Type: A

• Value: <INGRESS_EXTERNAL_IP>

• TTL: 300

DNS Propagation: DNS changes can take 5-15 minutes to propagate. Test with nslookup api-k8sagent.your-domain.com

---

Using Scoutflo-Managed Subdomain

Standard Domain

Use your own domain (recommended for production):

--set ingress.hostname=api-k8sagent.your-company.com \
--set ingress.tls.wildcardHostname=your-company.com

Scoutflo Subdomain

---

Troubleshooting Guide

Common Issues & Solutions

Pod Issues

Symptom	Possible Cause	Solution
ImagePullBackOff	Registry credentials invalid	Verify helm registry login was successful
CrashLoopBackOff	Configuration error	Check pod logs for auth/config issues
Pending state	Resource constraints	Check node resources and pod requests

Networking Issues

RBAC Issues

Getting Help

If you're still experiencing issues:

Contact Scoutflo Support with:

• Your Helm installation command (with secrets redacted)

• The diagnostic files above

• Your cluster information (provider, version, ingress controller)

Pro Tip: Most issues are resolved by verifying DNS, TLS certificates, or RBAC permissions. Double-check these first!

---

Next Steps

🎉 Congratulations! Your Scoutflo Kubernetes Agent is now monitoring your cluster and sending data to the Scoutflo platform.

Accessing Scoutflo

• Log into your Scoutflo dashboard to view cluster insights

• Configure alerting rules and notification channels

• Explore the AI-powered incident investigation features

---

Questions or feedback? Reach out to the Scoutflo team at [e-mail] or join our community Slack.