Skip to main content

​
Overview

This guide walks you through deploying a production-ready Kubernetes cluster on Google Kubernetes Engine (GKE). GKE provides a fully managed Kubernetes environment with automatic updates, built-in security, and seamless integration with Google Cloud services.

​
Prerequisites

Before you begin, ensure you have:
  • A Google Cloud Platform (GCP) account with billing enabled
  • gcloud CLI installed (Installation Guide)
  • kubectl installed (Installation Guide)
  • Appropriate IAM permissions to create GKE clusters

​
Initial Setup

​
1. Authenticate with Google Cloud

Log in to your Google Cloud account: 
gcloud auth login
This opens your browser to complete the authentication flow.

​
2. Initialize gcloud Configuration

Set up your default configuration: 
gcloud init
This interactive command will:
  • Select or create a configuration
  • Choose your Google Cloud project
  • Set default compute region and zone

​
3. Set Your Project

Switch to or confirm your target project: 
gcloud config set project [YOUR_PROJECT_ID]
Replace [YOUR_PROJECT_ID] with your actual GCP project ID.

​
Creating a GKE Cluster

​
Standard Cluster Creation

Create a standard GKE cluster using the GCP Console or gcloud CLI:

​
Using gcloud CLI

gcloud container clusters create exchange-cluster \
  --zone us-central1-a \
  --num-nodes 3 \
  --machine-type e2-standard-4 \
  --disk-size 50 \
  --enable-autoscaling \
  --min-nodes 3 \
  --max-nodes 10 \
  --enable-autorepair \
  --enable-autoupgrade
Production Best Practices:
  • Use at least 3 nodes for high availability
  • Enable autoscaling to handle traffic spikes
  • Enable auto-repair and auto-upgrade for maintenance
  • Choose appropriate machine types based on workload requirements

​
Using GCP Console

  1. Navigate to Kubernetes Engine > Clusters in the GCP Console
  2. Click Create Cluster
  3. Configure your cluster settings:
    • Name: Choose a descriptive name (e.g., exchange-cluster)
    • Location type: Zonal or Regional
    • Zone/Region: Select based on your users’ location
    • Node pool: Configure size, machine type, and autoscaling
GCP Console - Cluster Overview

​
View Cluster Status

List all clusters in your project: 
gcloud container clusters list
This displays:
  • Cluster name
  • Location
  • Master version
  • Node count
  • Status

​
Connecting to Your Cluster

​
Get Cluster Credentials

Configure kubectl to access your GKE cluster: 
gcloud container clusters get-credentials [CLUSTER_NAME] --location [LOCATION]
Example: 
gcloud container clusters get-credentials exchange-cluster --location us-central1-a
This command:
  • Fetches cluster endpoint and authentication data
  • Updates your kubeconfig file (~/.kube/config)
  • Sets the current context to your cluster

​
Verify Connection

Test your connection: 
kubectl cluster-info
kubectl get nodes

​
Managing Multiple Contexts

View all available contexts: 
kubectl config get-contexts
Switch between clusters: 
kubectl config use-context [CONTEXT_NAME]
Example: 
kubectl config use-context gke_my-project_us-central1-a_exchange-cluster

​
GCP Console Overview

The GCP Console provides powerful visualization and management tools for your GKE cluster.

​
Cluster Dashboard

GKE Cluster Dashboard
The dashboard shows:
  • Cluster health and status
  • Node pool information
  • Resource utilization
  • Cluster version and configuration

​
Workloads View

GKE Workloads
Monitor your deployments:
  • Running pods and their status
  • Deployment configurations
  • Resource requests and limits
  • Container logs and events

​
Services and Ingress

GKE Services
Manage networking:
  • Service endpoints
  • Load balancers
  • Ingress configurations
  • External IPs

​
Node Pools

GKE Node Pools
Configure compute resources:
  • Node pool size and autoscaling
  • Machine types and configurations
  • Node taints and labels
  • Upgrade strategies

​
Cluster Configuration

​
Enable Required APIs

Ensure necessary GCP APIs are enabled: 
gcloud services enable container.googleapis.com
gcloud services enable compute.googleapis.com

​
Configure Cluster Autoscaling

Update autoscaling settings: 
gcloud container clusters update [CLUSTER_NAME] \
  --enable-autoscaling \
  --min-nodes 3 \
  --max-nodes 10 \
  --location [LOCATION]

​
Resize Node Pool

Manually adjust node count: 
gcloud container clusters resize [CLUSTER_NAME] \
  --num-nodes 5 \
  --location [LOCATION]

​
Security Best Practices

Production Security Checklist:
  • Enable Workload Identity for secure GCP service access
  • Use private clusters to restrict public access
  • Enable Binary Authorization for container image verification
  • Configure network policies to control pod-to-pod communication
  • Regularly update to the latest GKE version
  • Use GCP IAM for fine-grained access control

​
Enable Workload Identity

gcloud container clusters update [CLUSTER_NAME] \
  --workload-pool=[PROJECT_ID].svc.id.goog \
  --location [LOCATION]

​
Monitoring and Logging

GKE integrates with Google Cloud’s operations suite:
  • Cloud Monitoring: Real-time metrics and alerting
  • Cloud Logging: Centralized log management
  • Cloud Trace: Distributed tracing for microservices
Access logs from kubectl: 
kubectl logs -f deployment/[DEPLOYMENT_NAME]

​
Upgrading Your Cluster

​
Check Available Versions

gcloud container get-server-config --location [LOCATION]

​
Upgrade Control Plane

gcloud container clusters upgrade [CLUSTER_NAME] \
  --master \
  --cluster-version [VERSION] \
  --location [LOCATION]

​
Upgrade Node Pools

gcloud container clusters upgrade [CLUSTER_NAME] \
  --node-pool [POOL_NAME] \
  --location [LOCATION]

​
Cost Optimization

Cost-Saving Strategies:
  • Use Preemptible VMs for non-critical workloads (up to 80% savings)
  • Enable cluster autoscaling to match demand
  • Use appropriate machine types (e2 series for general workloads)
  • Set resource requests and limits on pods
  • Delete unused load balancers and persistent disks

​
Troubleshooting

​
Common Issues

Cannot connect to cluster: 
# Re-fetch credentials
gcloud container clusters get-credentials [CLUSTER_NAME] --location [LOCATION]

# Verify context
kubectl config current-context
Insufficient quota errors: Check your project quotas in GCP Console under IAM & Admin > Quotas. Node not ready: 
kubectl describe node [NODE_NAME]
kubectl get events --sort-by=.metadata.creationTimestamp

​
Next Steps

ArgoCD GitOps

Set up continuous deployment with ArgoCD

gcloud CLI Reference

Master essential gcloud commands

​
Additional Resources

Build docs developers (and LLMs) love

Get started for freeTalk to us