Skip to content

Lab 3: Deploy Chaos Engine Victim

In this lab we are going to deploy test Kubernetes cluster in GCP with 3 worker nodes and 2 dummy applications. The cluster will be a target for our experiments we will run in the next lab.

Create new Kubernetes cluster

Let's provision tiny cluster with 3 nodes. Each node will have 1 CPU and 1.5 GB memory. From Cloud Shell run:

gcloud container clusters create "chaos-engine-victim" \
    --zone "europe-west1-b" \
    --no-enable-basic-auth \
    --cluster-version "1.14.10-gke.17" \
    --machine-type "g1-small" \
    --image-type "COS" \
    --disk-type "pd-standard" \
    --disk-size "20" \
    --num-nodes "3" \
    --enable-stackdriver-kubernetes \
    --enable-ip-alias \
    --enable-autoupgrade \
    --enable-autorepair 

WARNING: Starting in 1.12, default node pools in new clusters will have their legacy Compute Engine instance metadata endpoints disabled by default. To create a cluster with legacy instance metadata endpoints disabled in the default node pool, run `clusters create` with the flag `--metadata disable-legacy-endpoints=true`.
WARNING: The Pod address range limits the maximum size of the cluster. Please refer to https://cloud.google.com/kubernetes-engine/docs/how-to/flexible-pod-cidr to learn how to optimize IP address allocation.
This will enable the autorepair feature for nodes. Please see https://cloud.google.com/kubernetes-engine/docs/node-auto-repair for more information on node autorepairs.
Creating cluster chaos-engine-victim in europe-west1-b... Cluster is being health-checked (master is healthy)...done.
Created [https://container.googleapis.com/v1/projects/xxxx/zones/europe-west1-b/clusters/chaos-engine-victim].
To inspect the contents of your cluster, go to: https://console.cloud.google.com/kubernetes/workload_/gcloud/europe-west1-b/chaos-engine-victim?project=xxxx
kubeconfig entry generated for chaos-engine-victim.
NAME                 LOCATION        MASTER_VERSION  MASTER_IP      MACHINE_TYPE  NODE_VERSION    NUM_NODES  STATUS
chaos-engine-victim  europe-west1-b  1.14.10-gke.17  xxx.xxx.xxx.xxx  g1-small      1.14.10-gke.17  3          RUNNING

Authenticate

In order to run kubectl commands we need to add new context into your .kube config. It could be done by following gcloud command.

gcloud container clusters get-credentials chaos-engine-victim --zone europe-west1-b 

Fetching cluster endpoint and auth data.
kubeconfig entry generated for chaos-engine-victim.

Check that the kubectl context has been switched.

kubectl config current-context

xxxxxxxxx-west1-b_chaos-engine-victim

Deploy dummy applications

In order to demonstrate Chaos Engine features we need to deploy couple of applications. We will use 2 deployments, one nginx and second apache. Please go to the workshop repo you cloned to your Cloud Shell instance and perform following command.

kubectl apply -f kubernetes/applications.yml

deployment.apps/nginx configured
deployment.apps/apache created

Verify test application were started.

kubectl get pods

NAME                      READY   STATUS    RESTARTS   AGE
apache-7c99b8d54f-bkk2w   1/1     Running   0          77m
apache-7c99b8d54f-g8k6g   1/1     Running   0          77m
apache-7c99b8d54f-ncwcn   1/1     Running   0          77m
nginx-8779fd9dc-22hqf     1/1     Running   0          76m
nginx-8779fd9dc-pt66s     1/1     Running   0          77m
nginx-8779fd9dc-zn94b     1/1     Running   0          76m

Configure RBAC

In the last step we need to create a service account that will be used by the Engine and do a role bindings. All can be do by simply applying a template from the workshop repo.

kubectl apply -f kubernetes/rbac.yml

role.rbac.authorization.k8s.io/chaos-engine-role created
serviceaccount/chaos-engine-serviceaccount created
rolebinding.rbac.authorization.k8s.io/chaos-engine-rolebinding created

Lab summary

At the end of this exercise you should have:

  • Running Kubernetes cluster
  • Deployed dummy applications
  • RBAC configured