/getting-to-know-k8s

Permalink
Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
Raw Blame History
453 lines (299 sloc) 20.1 KB

Using Scripts

In addition to the step-by-step instructions provided for each lab, this repository also contains scripts to automate some of the activities being performed in the labs.

You will need to review the script variables guide before running any of the provided scripts.

If you are using the Vagrant box together with Windows & Git Bash, you may run into a issue where the script files are not marked as executable due to the underlying Windows file system. The easiest workaround for this issue is to clone the repository to a different folder from within the Vagrant box and you should then be able to run the scripts there:

git clone https://github.com/kumorilabs/getting-to-know-k8s ~/getting-to-know-k8s-scripts
cd ~/getting-to-know-k8s-scripts

Cluster Deployment

Cluster Add-ons

Federation

Application Deployment

Cluster Deployment

Create Cluster

Usage

./scripts/clusters/create-cluster.sh [CLUSTER_ALIAS] [CLUSTER_AWS_AZ]
  • [CLUSTER_ALIAS] - Cluster context alias to use for the new cluster (Optional)
  • [CLUSTER_AWS_AZ] - AWS availability zone where the new cluster will be deployed (Optional)

If you do provide the configuration options above for [CLUSTER_ALIAS] and [CLUSTER_AWS_AZ], they will override the values specified in the variables.sh file.

Example

The following will create two clusters, one with the values from the variables.sh file (default is usa & us-east-1a) and the other with a cluster alias of eur, which will be located in the eu-west-1a AWS availability zone:

./scripts/clusters/create-cluster.sh
./scripts/clusters/create-cluster.sh eur eu-west-1a

Delete Cluster

Usage

./scripts/clusters/delete-cluster.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

Example

The following will delete two clusters, one with the value from the variables.sh file (default is usa) and the other with cluster alias eur:

./scripts/clusters/delete-cluster.sh
./scripts/clusters/delete-cluster.sh eur

Cluster Add-ons

Deploy Kubernetes Dashboard

Usage

./scripts/addons/kubernetes-dashboard/deploy-dashboard.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

Example

The following will deploy the Kubernetes dashboard add-on to two clusters, usa (default in the variables.sh file) and eur:

./scripts/addons/kubernetes-dashboard/deploy-dashboard.sh
./scripts/addons/kubernetes-dashboard/deploy-dashboard.sh eur

Deploy Heapster Monitoring

Usage

./scripts/addons/heapster-monitoring/deploy-heapster-monitoring.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

Example

The following will deploy the Heapster monitoring add-on to two clusters, usa (default in the variables.sh file) and eur:

./scripts/addons/heapster-monitoring/deploy-heapster-monitoring.sh
./scripts/addons/heapster-monitoring/deploy-heapster-monitoring.sh eur

Deploy Cluster Autoscaler

Usage

./scripts/addons/cluster-autoscaler/deploy-cluster-autoscaler.sh [CLUSTER_ALIAS] [CLUSTER_AWS_REGION] [MIN_NODES] [MAX_NODES]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Mandatory)
  • [CLUSTER_AWS_REGION] - AWS region of the existing cluster (Mandatory)
  • [MIN_NODES] - Minimum number of nodes to configure for the AWS autoscaling group (Mandatory)
  • [MAX_NODES] - Maximum number of nodes to configure for the AWS autoscaling group (Mandatory)

Example

The following will deploy the cluster autoscaler add-on to two clusters, one with cluster alias usa located in the us-east-1 AWS region with a minimum of 3 nodes and a maximum of 5 nodes. And the other with cluster alias eur located in the eu-west-1 AWS region with a minimum of 4 nodes and a maximum of 6 nodes:

./scripts/addons/cluster-autoscaler/deploy-cluster-autoscaler.sh usa us-east-1 3 5
./scripts/addons/cluster-autoscaler/deploy-cluster-autoscaler.sh eur eu-west-1 4 6

Federation

Create Federation

Usage

./scripts/clusters/federation/create-federation.sh [HOST_CLUSTER_ALIAS] [CLUSTER_2_ALIAS] [CLUSTER_3_ALIAS]
  • [HOST_CLUSTER_ALIAS] - Cluster context alias of an existing cluster, which will host the Federation services (Mandatory)
  • [CLUSTER_2_ALIAS] - Cluster context alias of a second existing cluster, which will join the Federation with the host cluster (Mandatory)
  • [CLUSTER_3_ALIAS] - Cluster context alias of a third existing cluster, which will join the Federation with the host cluster (Optional)

Example

The following will create a Federation between two clusters, usa & eur. The usa cluster will host the Federation services:

./scripts/clusters/federation/create-federation.sh usa eur

The following will create a Federation between three clusters, frankfurt, sydney & mumbai. The frankfurt cluster will host the Federation services:

./scripts/clusters/federation/create-federation.sh frankfurt sydney mumbai

Application Deployments

Deploy Hugo Site (Standalone)

Usage

./scripts/apps/hugo-app/deploy-hugo-app.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

Example

The following will deploy the Hugo site to two clusters, usa (default in the variables.sh file) and eur:

./scripts/apps/hugo-app/deploy-hugo-app.sh
./scripts/apps/hugo-app/deploy-hugo-app.sh eur

Deploy Hugo Site (Federated)

Usage

./scripts/apps/hugo-app-federation/deploy-hugo-app-federation.sh [CLUSTER_1_ALIAS] [CLUSTER_1_AWS_REGION] [CLUSTER_2_ALIAS] [CLUSTER_2_AWS_REGION] [CLUSTER_3_ALIAS] [CLUSTER_3_AWS_REGION]
  • [CLUSTER_1_ALIAS] - Cluster context alias of the first existing cluster in a Federation (Mandatory)
  • [CLUSTER_1_AWS_REGION] - AWS region of the first existing cluster in a Federation (Mandatory)
  • [CLUSTER_2_ALIAS] - Cluster context alias of the second existing cluster in a Federation (Mandatory)
  • [CLUSTER_2_AWS_REGION] - AWS region of the second existing cluster in a Federation (Mandatory)
  • [CLUSTER_3_ALIAS] - Cluster context alias of the third existing cluster in a Federation (Optional)
  • [CLUSTER_3_AWS_REGION] - AWS region of the third existing cluster in a Federation (Optional)

Example

The following will deploy the federated Hugo site to three clusters, usa in the us-east-1 AWS region, eur in the eu-west-1 AWS region and tyo in the ap-northeast-1 AWS region:

./scripts/apps/hugo-app-federation/deploy-hugo-app-federation.sh usa us-east-1 eur eu-west-1 tyo ap-northeast-1

Deploy Hugo Site (Rolling Update)

  • Creates a rolling update Deployment scenario of the Hugo site in a single cluster
  • Docker image of the Hugo site is specified as HUGO_APP_DOCKER_IMAGE in the /scripts/variables.sh file
  • Docker image tags red & yellow will be used
  • Creates a load balancer Service for the Deployment, which exposes port 80
  • Creates a DNS record with the prefix hugo-rolling-update in your domain in Route 53 (ex: hugo-rolling-update.k8s.kumorilabs.com)
  • Link to Rolling Updates in Lab #4: Deployment Strategies: Rolling Updates, Canary & Blue-Green

Usage

./scripts/apps/hugo-app-rolling-update/deploy-hugo-app-rolling-update.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

The initial Deployment will be using the red image. To kick-off the rolling update to the yellow image, run the following:

# Set your GitHub username
export GITHUB_USERNAME="smesch"
kubectl set image deploy/hugo-app-rolling-update \
        hugo-app=${DOCKER_HUB_USERNAME}/hugo-app:yellow

Example

The following will deploy the Hugo site to two clusters, usa (default in the variables.sh file) and eur:

./scripts/apps/hugo-app-rolling-update/deploy-hugo-app-rolling-update.sh
./scripts/apps/hugo-app-rolling-update/deploy-hugo-app-rolling-update.sh eur

Deploy Hugo Site (Blue-Green Deployment)

Usage

./scripts/apps/hugo-app-blue-green/deploy-hugo-app-blue-green.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

The initial Service will be using the color: blue selector. To change the selector to color: green, run the following:

kubectl set selector svc/hugo-app-blue-green-svc color=green

Example

The following will deploy the Hugo site to two clusters, usa (default in the variables.sh file) and eur:

./scripts/apps/hugo-app-blue-green/deploy-hugo-app-blue-green.sh
./scripts/apps/hugo-app-blue-green/deploy-hugo-app-blue-green.sh eur

Deploy Hugo Site (Canary Deployment)

Usage

./scripts/apps/hugo-app-canary/deploy-hugo-app-canary.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

The initial Deployments will be three replicas of the red image and one replica of the yellow image. To change the amount of replicas for the two Deployments, run the following:

kubectl scale deployment hugo-app-yellow --replicas=4
kubectl scale deployment hugo-app-red --replicas=2

Example

The following will deploy the Hugo site to two clusters, usa (default in the variables.sh file) and eur:

./scripts/apps/hugo-app-canary/deploy-hugo-app-canary.sh
./scripts/apps/hugo-app-canary/deploy-hugo-app-canary.sh eur

Deploy Jenkins

  • Creates a Deployment of Jenkins in a single cluster
  • Docker image of Jenkins is specified as JENKINS_DOCKER_IMAGE in the /scripts/variables.sh file
  • Creates a persistent volume claim for the Jenkins home directory (/var/jenkins_home)
  • Creates a load balancer Service for the Jenkins Deployment, which exposes ports 80
  • Creates a DNS record with the prefix jenkins in your domain in Route 53 (ex: jenkins.k8s.kumorilabs.com)
  • Link to Deploy and configure Jenkins in Lab #6: Integrating Jenkins and Kubernetes

Usage

./scripts/apps/jenkins/deploy-jenkins.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

Example

The following will deploy Jenkins to two clusters, usa (default in the variables.sh file) and eur:

./scripts/apps/jenkins/deploy-jenkins.sh
./scripts/apps/jenkins/deploy-jenkins.sh eur

Deploy Horizontal Pod Autoscaling Demo

Usage

./scripts/apps/horizontal-pod-autoscaling/deploy-hpa-app.sh [CLUSTER_ALIAS]
  • [CLUSTER_ALIAS] - Cluster context alias of an existing cluster (Optional)

If you do provide the configuration option above for [CLUSTER_ALIAS], it will override the value specified in the variables.sh file.

After running the script, open up a second terminal (if using Vagrant, browse to the root of the repository from a second terminal and connect with vagrant ssh) and run the following:

kubectl run -i --tty load-generator --image=busybox /bin/sh
# Hit enter for command prompt
while true; do wget -q -O- http://php-apache.default.svc.cluster.local; done
OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!
OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!
OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!OK!...

Example

The following will deploy the horizontal pod autoscaling demo to two clusters, usa (default in the variables.sh file) and eur:

./scripts/apps/horizontal-pod-autoscaling/deploy-hpa-app.sh
./scripts/apps/horizontal-pod-autoscaling/deploy-hpa-app.sh eur