GKE Cheatsheet

This document was migrated from DigiDocs

In this section, you will find my notes on using Kubernetes on Google Cloud Platform's Kubernetes Engine (GKE). It covers a typical workflow for starting a Kubernetes cluster and deploying an application on it.

Clusters#

Creating a cluster#

Typical command:

gcloud container clusters create $C_NAME --zone $ZONE --num-nodes $NUM

Shorter command:

gcloud container clusters create $C_NAME

With auto-scaling

gcloud container clusters create $C_NAME --enable-autoscaling --min-nodes $MIN --max-nodes $MAX
note

The default value for --num-nodes is 3

tip

You can set the default zone for Kubernetes Engine using the following:

  • For zonal clusters: gcloud config set compute/zone $ZONE
  • For regional clusters: gcloud config set compute/region $REGION

Deleting a cluster#

Typical command

gcloud container clusters delete $C_NAME

Getting cluster info#

Get all pods running in the cluster

kubectl get pods

Get all services running in the cluster

kubectl get services

Get credentials for cluster

gcloud container clusters get-credentials $C_NAME
Cite

This command enables switching to a specific cluster, when working with multiple clusters. It can also be used to access a previously created cluster from a new workstation.

Deployments#

Creating a deployment#

To create a deployment, you need to have your Docker image prepared beforehand. This image must be built and uploaded to the Container Registry before you can deploy it on your GKE cluster.

tip

Before you proceed, you need to configure Docker to authenticate to the Container Registry: gcloud auth configure-docker

  1. Build your image:
    docker build -t gcr.io/$PROJECT_ID/$NAME:$VER .
  2. Verify it was built:
    docker images
  3. Upload your image to the registry:
    docker push gcr.io/$PROJECT_ID/$NAME:$VER
  4. Verify it was uploaded:
    docker run --rm -p $CONT_PORT:$HOST_PORT gcr.io/$PROJECT_ID/$NAME:$VER
  5. Create your deployment:
    kubectl create deployment $D_NAME --image=gcr.io/$PROJECT_ID/$NAME:$VER
  6. Verify it was deployed:
    kubectl get pods
  7. Expose the deployment to the Internet via a Service resource:
    kubectl expose deployment $D_NAME --type=LoadBalancer --port $EXPOSED_PORT --target-port $HOST_PORT
  8. Verify the service is running:
    kubectl get service

Deleting a deployment#

  1. Delete the Service resource:
    kubectl delete service $D_NAME
  2. Delete the cluster:
    gcloud container clusters delete $C_NAME

Updating your deployment#

To update your deployment with a new Docker image, you have to upload it to the Cloud Registry. Next, you can apply a rolling update of your deployment's Docker image.

  1. Build your new image (remember to update $VER):
    docker build -t gcr.io/$PROJECT_ID/$I_NAME:$VER .
  2. Verify it was built:
    docker images
  3. Upload your image to the registry:
    docker push gcr.io/$PROJECT_ID/$I_NAME:$VER
  4. Verify it was uploaded:
    docker run --rm -p $CONT_PORT:$HOST_PORT gcr.io/$PROJECT_ID/$I_NAME:$VER
  5. Apply a rolling image update:
    kubectl set image deployment/$D_NAME $I_NAME=gcr.io/$PROJECT_ID/$I_NAME:$VER

Scaling a deployment#

Typical example

kubectl scale deployment $D_NAME --replicas $NUM

With auto-scaling

kubectl autoscale deployment $D_NAME --max $MAX --min $MIN --cpu-percent $PERCENT

Getting deployment info#

Typical example

kubectl get deployment $D_NAME

Resources#

Edit this page