Why managed Kubernetes matters in GCP - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a GKE cluster using gcloud CLI.

GCP

gcloud container clusters create [1] --zone us-central1-a

Drag options to blanks, or click blank then click option'

Acluster-1

Bdefault

Ccreate-cluster

Dmy-cluster

Attempts:

3 left

💡 Hint

Common Mistakes

Using a command word instead of a cluster name

Leaving the cluster name blank

The command requires a cluster name after 'create'. 'my-cluster' is a valid name.

2fill in blank

medium

Complete the command to get credentials for your GKE cluster.

GCP

gcloud container clusters get-credentials [1] --zone us-central1-a

Drag options to blanks, or click blank then click option'

Amy-cluster

Btest-cluster

Cdefault

Dcluster-2

Attempts:

3 left

💡 Hint

Common Mistakes

Using a wrong cluster name

Omitting the cluster name

You must specify the exact cluster name to get its credentials. 'my-cluster' matches the created cluster.

3fill in blank

hard

Fix the error in the command to list nodes in your GKE cluster.

GCP

kubectl get [1]

Drag options to blanks, or click blank then click option'

Anodes

Bpods

Cservices

Ddeployments

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'pods' instead of 'nodes'

Using 'services' or 'deployments' which list different resources

To see the nodes in the cluster, use 'kubectl get nodes'.

4fill in blank

hard

Fill both blanks to create a deployment with 3 replicas of nginx.

GCP

kubectl create deployment [1] --image=[2] --replicas=3

Drag options to blanks, or click blank then click option'

Anginx-deploy

Bnginx:latest

Cnginx

Dnginx-image

Attempts:

3 left

💡 Hint

Common Mistakes

Using image name without tag

Using deployment name as image

The deployment name can be 'nginx-deploy' and the image should be 'nginx:latest' to pull the official nginx image.

5fill in blank

hard

Fill all three blanks to expose the nginx deployment as a LoadBalancer service on port 80.

GCP

kubectl expose deployment [1] --type=[2] --port=[3]

Drag options to blanks, or click blank then click option'

Anginx-deploy

BLoadBalancer

C80

DClusterIP

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'ClusterIP' instead of 'LoadBalancer'

Wrong port number

Expose the 'nginx-deploy' deployment as a 'LoadBalancer' type service on port 80 to allow external traffic.

Practice

1. What is the main benefit of using managed Kubernetes services like Google Kubernetes Engine (GKE)?

easy

A. It only works with Google Cloud Storage.

B. It requires you to manually update and secure the cluster nodes.

C. It handles infrastructure management so you can focus on your applications.

D. It eliminates the need to write application code.

Solution

1. Step 1: Understand managed Kubernetes purpose

   Managed Kubernetes services automate infrastructure tasks like updates and scaling.

2. Step 2: Focus on application development

   This automation lets developers focus on building and deploying apps instead of managing servers.

3. Final Answer:

   It handles infrastructure management so you can focus on your applications. -> Option C

4. Quick Check:

   Managed Kubernetes = Focus on apps [OK]

Hint: Managed Kubernetes means less server work, more app focus [OK]

Common Mistakes:

• Thinking you must manage updates manually
• Believing it only works with specific storage
• Assuming it removes the need to code

2. Which command correctly creates a new Kubernetes cluster using Google Kubernetes Engine (GKE)?

easy

A. gcloud container clusters create my-cluster --zone us-central1-a

B. kubectl create cluster my-cluster --region us-central1

C. gcloud compute clusters new my-cluster --zone us-central1-a

D. kubectl cluster create my-cluster --zone us-central1-a

Solution

1. Step 1: Identify correct GKE cluster creation command

   The correct command uses 'gcloud container clusters create' with cluster name and zone.

2. Step 2: Check syntax and tool usage

   'kubectl' manages resources inside clusters, not cluster creation; 'gcloud compute' is for VM instances.

3. Final Answer:

   gcloud container clusters create my-cluster --zone us-central1-a -> Option A

4. Quick Check:

   Cluster creation uses gcloud container clusters create [OK]

Hint: Use 'gcloud container clusters create' to make GKE clusters [OK]

Common Mistakes:

• Using kubectl to create clusters
• Confusing compute commands with container commands
• Missing required flags like --zone

3. What will be the output of the following command when run on a managed Kubernetes cluster?

kubectl get nodes

medium

A. Lists all nodes in the cluster with their status and roles.

B. Creates a new node in the cluster.

C. Deletes all nodes in the cluster.

D. Shows the logs of the nodes.

Solution

1. Step 1: Understand 'kubectl get nodes' command

   This command lists all nodes in the Kubernetes cluster, showing their status and roles.

2. Step 2: Differentiate from other commands

   It does not create, delete, or show logs; those require different commands.

3. Final Answer:

   Lists all nodes in the cluster with their status and roles. -> Option A

4. Quick Check:

   'kubectl get nodes' = list nodes [OK]

Hint: 'kubectl get nodes' shows cluster nodes status [OK]

Common Mistakes:

• Thinking it creates or deletes nodes
• Confusing it with log commands
• Expecting detailed pod info instead

4. You tried to create a GKE cluster with this command:

gcloud container clusters create my-cluster --zone us-central1-a --num-nodes -3

What is the problem and how to fix it?

medium

A. The zone flag is incorrect; use --region instead.

B. The number of nodes cannot be negative; use a positive integer instead.

C. The cluster name is invalid; it must start with a capital letter.

D. The command is missing the --machine-type flag.

Solution

1. Step 1: Identify invalid flag value

   The --num-nodes flag cannot be negative; '-3' is invalid.

2. Step 2: Correct the flag value

   Replace '-3' with a positive integer like '3' to specify node count.

3. Final Answer:

   The number of nodes cannot be negative; use a positive integer instead. -> Option B

4. Quick Check:

   Node count must be positive [OK]

Hint: Node count must be positive number, not negative [OK]

Common Mistakes:

• Using negative numbers for node count
• Confusing zone and region flags
• Thinking cluster names must be capitalized

5. You want to ensure your managed Kubernetes cluster automatically updates and scales without downtime. Which feature of managed Kubernetes helps achieve this?

hard

A. Disabling auto-updates to control changes.

B. Manual patching and fixed node counts.

C. Using only pre-built container images.

D. Automatic node upgrades and cluster autoscaling.

Solution

1. Step 1: Identify features for zero downtime

   Managed Kubernetes offers automatic node upgrades and autoscaling to keep clusters updated and responsive.

2. Step 2: Understand why other options fail

   Manual patching risks downtime; disabling updates stops improvements; pre-built images don't handle scaling or upgrades.

3. Final Answer:

   Automatic node upgrades and cluster autoscaling. -> Option D

4. Quick Check:

   Auto upgrades + autoscaling = smooth updates and scaling [OK]

Hint: Auto upgrades and autoscaling keep clusters healthy [OK]

Common Mistakes:

• Thinking manual patching avoids downtime
• Disabling auto-updates for control
• Believing container images handle scaling