esddns

ArgoCD Deployment Guide for ESDDNS Operator

This guide provides comprehensive instructions for deploying the ESDDNS Operator using ArgoCD with both Helm and Kustomize options.

Architecture Overview

ArgoCD (in argocd namespace)
├── AppProject: esddns
├── Application: esddns-operator-helm       → Helm deployment
├── Application: esddns-operator-kustomize  → Kustomize deployment
├── Application: esddns-operator-helm-dev   → Development environment
└── Application: esddns-operator-helm-prod  → Production environment

Quick Start

1. Prerequisites

• Kubernetes cluster (1.20+)
• kubectl configured
• helm (optional, for manual Helm operations)
• ArgoCD v2.4+ installed

2. Install ArgoCD (if not already installed)

# Create namespace
kubectl create namespace argocd

# Install ArgoCD
helm repo add argo https://argoproj.github.io/argo-helm
helm repo update
helm install argocd argo/argo-cd -n argocd \
  --set configs.params."application.instanceLabelKey"="argocd.argoproj.io/instance"

3. Deploy ESDDNS Applications

Option A: Deploy with Bootstrap Script

cd argocd/bootstrap
./argocd-setup.sh

Option B: Manual Deployment

# Deploy AppProject and all applications
kubectl apply -k argocd/

# Or deploy specific applications
kubectl apply -f argocd/argocd-appproject.yaml
kubectl apply -f argocd/applications/esddns-operator-helm.yaml

Option C: Deploy with Environment Overlay

# For development
kubectl apply -k argocd/overlays/dev/

# For production
kubectl apply -k argocd/overlays/prod/

Deployment Strategies

Strategy 1: Helm-Only Deployment

Use this if you prefer Helm for all deployments.

kubectl apply -f argocd/applications/esddns-operator-helm.yaml

# Monitor the application
argocd app get esddns-operator-helm
argocd app wait esddns-operator-helm --sync

Pros:

• Simpler, single approach
• Direct values override capability
• Environment-specific values files

Cons:

• Less flexibility for complex overlays
• Limited patch capability

Strategy 2: Kustomize-Only Deployment

Use this if you have complex overlays and patches.

kubectl apply -f argocd/applications/esddns-operator-kustomize.yaml

# Monitor the application
argocd app get esddns-operator-kustomize
argocd app wait esddns-operator-kustomize --sync

Pros:

• Powerful patching and overlays
• Strong community support
• Excellent for multi-environment setups

Cons:

• Learning curve for complex setups
• More files to manage

Strategy 3: Dual Deployment (Dev + Prod)

Deploy different versions to different environments.

# Production with Helm
kubectl apply -f argocd/applications/esddns-operator-helm-prod.yaml

# Development with Kustomize
kubectl apply -f argocd/applications/esddns-operator-kustomize.yaml

# Monitor both
argocd app get esddns-operator-helm-prod
argocd app get esddns-operator-kustomize

Configuration Management

Helm Configuration

Update values in helm/esddns-operator/values.yaml:

replicaCount: 2
image:
  repository: sqe/esddns
  tag: latest
resources:
  limits:
    cpu: 500m
    memory: 512Mi

For environment-specific values, use:

• values-development.yaml - Development overrides
• values-production.yaml - Production overrides

Kustomize Configuration

Update k8s/overlays/production/kustomization.yaml:

resources:
  - ../../base

images:
  - name: esddns-operator
    newTag: v1.0.0

replicas:
  - name: esddns-operator-service
    count: 3

commonLabels:
  environment: production

Sync Policies

Manual Sync (Default - Safe for Production)

syncPolicy:
  syncOptions:
    - CreateNamespace=true
  automated: null
  retry:
    limit: 5

Manual sync keeps you in control:

# Sync the application
argocd app sync esddns-operator-helm

# Sync with prune (delete removed resources)
argocd app sync esddns-operator-helm --prune

# Force sync
argocd app sync esddns-operator-helm --force

Automatic Sync (Development Only)

syncPolicy:
  automated:
    prune: true
    selfHeal: true

Once enabled, every Git commit automatically deploys to the cluster.

Selective Sync

Sync only specific resources:

argocd app sync esddns-operator-helm --resource apps:Deployment:esddns-operator-service

Monitoring and Troubleshooting

Check Application Status

# Get application status
kubectl get application -n argocd
kubectl get application esddns-operator-helm -n argocd -o yaml

# Using ArgoCD CLI
argocd app list
argocd app get esddns-operator-helm
argocd app logs esddns-operator-helm

View Deployment Status

# Check deployed resources
kubectl get all -n esddns-operator
kubectl describe deployment esddns-operator-service -n esddns-operator

# Check for errors
kubectl logs -n esddns-operator -l app=esddns-operator-service

Common Issues

Application stuck in “Progressing”

# Sync manually
argocd app sync esddns-operator-helm --force

# Check resource conditions
kubectl describe pod -n esddns-operator

Resources not being created

# Check AppProject permissions
kubectl get appproject esddns -n argocd -o yaml

# Check application source is valid
argocd app get esddns-operator-helm --refresh

Sync failure

# Get detailed error information
argocd app get esddns-operator-helm --hard-refresh
kubectl describe application esddns-operator-helm -n argocd

# Check Git repository access
argocd repo list

Access ArgoCD UI

Port Forward

kubectl port-forward svc/argocd-server -n argocd 8080:443
# Open: https://localhost:8080

Get Admin Password

kubectl -n argocd get secret argocd-initial-admin-secret -o jsonpath='{.data.password}' | base64 -d

Using Ingress (Optional)

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: argocd-server
  namespace: argocd
spec:
  ingressClassName: nginx
  rules:
    - host: argocd.example.com
      http:
        paths:
          - path: /
            pathType: Prefix
            backend:
              service:
                name: argocd-server
                port:
                  number: 443

Advanced Usage

GitOps Workflow

1. Make changes to Git repository
2. Create pull request with updates
3. ArgoCD detects changes
4. Review in ArgoCD UI
5. Merge PR
6. ArgoCD automatically syncs (if auto-sync enabled)

Multiple Environments

Deploy to different clusters:

# For cluster 1 (dev)
kubectl apply -k argocd/overlays/dev/ --context dev-cluster

# For cluster 2 (prod)
kubectl apply -k argocd/overlays/prod/ --context prod-cluster

Automated Upgrades

Enable automatic image tag updates with ArgoCD Image Updater:

helm install argocd-image-updater argo/argocd-image-updater -n argocd

Update Application with annotation:

metadata:
  annotations:
    argocd-image-updater.argoproj.io/image-list: esddns=sqe/esddns
    argocd-image-updater.argoproj.io/esddns.update-strategy: semver~1.0
    argocd-image-updater.argoproj.io/esddns.pull-secret: github

Notifications

Setup Slack/Teams notifications for sync events:

# Install ArgoCD Notifications
helm install argocd-notifications argo/argocd-notifications -n argocd

Cleanup

Remove Specific Application

kubectl delete application esddns-operator-helm -n argocd

Remove All ESDDNS Applications

kubectl delete -k argocd/

Full Cleanup (including ArgoCD)

kubectl delete namespace argocd
kubectl delete namespace esddns-operator

Best Practices

1. Use Kustomize for Complex Overlays - Easier to manage multiple environments
2. Keep Git as Source of Truth - All changes through Git
3. Enable Automatic Sync for Non-Prod - Manual sync for production
4. Use AppProject for Multi-Team - Control access and permissions
5. Monitor Application Health - Set up notifications and alerts
6. Version Control All Manifests - Track changes in Git
7. Use Semantic Versioning - Tag releases consistently
8. Test in Dev First - Validate changes before production

References

• ArgoCD Documentation
• Helm Documentation
• Kustomize Documentation
• ESDDNS Repository

This site is open source. Improve this page.