Google Kubernetes Engine (GKE) - A Complete Practical Guide

Explore updated GKE screen options and a simplified cluster creation flow with new menu sections, noting that the standard and enterprise tier choice is no longer shown.

Create a standard cluster in GKE using the UI or gcloud, then set up regional multi-zone deployment. Select the regular release channel and a stable version, such as 1.27.

Discover how GKE now uses a single enable private nodes checkbox to configure private versus public clusters, with private nodes lacking public IPs and visibility outside the VPC.

Connect to a Kubernetes cluster from a local laptop using gcloud auth login and the auth plugin, authenticating with your Gmail ID. Verify with kubectl by creating a test pod.

Connect to your GKE cluster from cloud shell in the browser, authenticate with your Gmail ID, and verify the test pod by running kubectl get pods.

Enable kubectl autocomplete and shortcuts in a GKE environment by configuring aliases in your profile, enabling tab completion, and using quick letters like k, p, and s for common tasks.

Explore pods as containers wrapped by Kubernetes with support for multiple containers and sidecar patterns, and learn deployments that manage pod replicas, replica sets, and declarative YAML with kubectl workflows.

Create and expose a front end deployment and service in GKE, using environment variables, a back end service DNS, and a load balancer to connect front end to back end.

Allocate and reserve an external static IP in GCP, then bind it to your load balancer service to keep a stable address even if you recreate the service.

Learn to connect a Google Kubernetes Engine cluster to a Cloud SQL instance, including public and private access via Google Private Access, for your back-end application.

Learn workload identity to securely connect pods to Cloud SQL without sharing json keys by linking a Kubernetes service account to a gcp service account and using iam based authentication.

Set up the workload identity by creating a GCP service account with SQL admin access, a Kubernetes service account, and linking and annotating them for seamless Cloud SQL access.

Implement sidecar pattern by running the SQL proxy with workload identity and auth proxy to connect the backend to a private SQL instance via localhost.

Explore Kubernetes volumes in GKE, from ephemeral pod storage to external persistent disks. See how PV, PVC, storage class, and mount paths enable dynamic provisioning and data persistence.

Configure a regional GKE cluster with two zones, one node per zone, and a 30 GB boot disk to define host volumes and enable file store and cloud storage drivers.

Demonstrates that existing persistent disks must be in the same zone as the GKE cluster; a zone mismatch prevents PV/PVC binding and leaves pods pending.

Define a custom storage class in Kubernetes to tailor reclaim policy and binding mode, provisioning a PD SSD immediately, then bind it to a PVC and run the pod.

Create a file share in GKE by using an existing storage class and PVC, explaining RWX file stores, daemonset pods, and the provisioning delay until the first consumer.

Compare storage options for GKE volumes by examining persistent disk, file store, and bucket, highlighting zonal constraints, read-write access, pricing, and use cases.

Learn how Kubernetes services expose apps inside the cluster and to the internet, focusing on cluster IP, NodePort, and load balancer, with front end and back end port considerations.

Deploy a three-replica Kubernetes deployment and expose it with a NodePort service, test access from cluster, VPC, and internet with firewall rules.

Explore how to expose apps outside the cluster using load balancer type services, ingress, and gateway in GKE, with internal and external options and TCP pass-through details.