GKE networking (VPC-native) in GCP - Mini Project: Build & Apply

GKE Networking with VPC-native Clusters

📖 Scenario: You are setting up a Google Kubernetes Engine (GKE) cluster that uses VPC-native networking. This means the cluster will use alias IP ranges from a Virtual Private Cloud (VPC) network for pod IPs, improving network management and scalability.

🎯 Goal: Build a VPC-native GKE cluster configuration step-by-step. You will first define the network and subnetwork, then configure the IP ranges, create the cluster with VPC-native settings, and finally enable the cluster's IP aliasing feature.

📋 What You'll Learn

Create a VPC network named gke-vpc-network

Create a subnetwork named gke-subnet in region us-central1 with IP range 10.0.0.0/16

Define secondary IP ranges for pods and services

Create a GKE cluster named vpc-native-cluster using the VPC-native IP aliasing

Enable IP aliasing in the cluster configuration

💡 Why This Matters

🌍 Real World

VPC-native GKE clusters improve network scalability and management by assigning IP addresses to pods from VPC secondary ranges, avoiding IP conflicts and simplifying routing.

💼 Career

Understanding VPC-native GKE networking is essential for cloud engineers and DevOps professionals managing scalable Kubernetes clusters on Google Cloud.

1

Create the VPC network and subnetwork

Create a VPC network called gke-vpc-network with autoCreateSubnetworks set to false. Then create a subnetwork called gke-subnet in region us-central1 with IP range 10.0.0.0/16.

GCP

# Create VPC network 'gke-vpc-network' with autoCreateSubnetworks false
# Create subnetwork 'gke-subnet' in 'us-central1' with IP range '10.0.0.0/16'

Need a hint?

Use google_compute_network resource for the VPC and google_compute_subnetwork for the subnet.

2

Define secondary IP ranges for pods and services

Add two secondary IP ranges to the gke-subnet: one named pods with IP range 10.4.0.0/14 and another named services with IP range 10.8.0.0/20.

GCP

resource "google_compute_network" "gke_vpc_network" {
  name                    = "gke-vpc-network"
  auto_create_subnetworks = false
}

resource "google_compute_subnetwork" "gke_subnet" {
  name          = "gke-subnet"
  ip_cidr_range = "10.0.0.0/16"
  region        = "us-central1"
  network       = google_compute_network.gke_vpc_network.id
  # Add secondary IP ranges for pods and services here
}

Need a hint?

Use secondary_ip_range blocks inside the google_compute_subnetwork resource.

3

Create the GKE cluster with VPC-native IP aliasing

Create a GKE cluster resource named vpc-native-cluster in region us-central1 that uses the gke-subnet and enables IP aliasing with the secondary ranges pods and services.

GCP

resource "google_container_cluster" "vpc_native_cluster" {
  name     = "vpc-native-cluster"
  location = "us-central1"
  network  = google_compute_network.gke_vpc_network.id
  subnetwork = google_compute_subnetwork.gke_subnet.name
  # Enable IP aliasing and configure cluster networking here
}

Need a hint?

Use the ip_allocation_policy block inside the cluster resource to enable IP aliasing.

4

Enable the cluster's IP aliasing feature

Add the final configuration line enable_ip_alias = true inside the google_container_cluster resource vpc_native_cluster to explicitly enable IP aliasing.

GCP

resource "google_container_cluster" "vpc_native_cluster" {
  name       = "vpc-native-cluster"
  location   = "us-central1"
  network    = google_compute_network.gke_vpc_network.id
  subnetwork = google_compute_subnetwork.gke_subnet.name

  ip_allocation_policy {
    use_ip_aliases                = true
    cluster_secondary_range_name  = "pods"
    services_secondary_range_name = "services"
  }
  # Add enable_ip_alias = true here
}

Need a hint?

Add enable_ip_alias = true inside the google_container_cluster resource.