Google Professional Cloud Network Engineer Question 331
Single answerGoogle Cloud PlatformYou are managing a GKE cluster using VPC-native (alias IPs) networking. The cluster has exhausted its current Pod IP range, and new Pods cannot be scheduled due to IP allocation issues. How can you resolve this problem while ensuring minimal service disruption?
- A
Resize the existing Pod IP range by modifying the subnet's CIDR to a larger range.
- B
Deploy an additional Pod range to the existing cluster using the
gcloudCLI. - C
Migrate the cluster to a new subnet with a larger IP range.
- D
Recreate the cluster with a larger Pod IP range and redeploy the workloads.
Show answer and explanation
Correct answer: B
Explanation
When a GKE cluster using VPC-native networking exhausts its Pod IP range, the best solution is to deploy an additional Pod range to the cluster. This can be done using the gcloud container clusters update command without requiring cluster recreation or downtime. Resizing existing ranges or migrating to new subnets is not supported or would cause significant disruption.
- A. Incorrect.
This is incorrect because you cannot resize an existing subnet’s allocated Pod IP range for a GKE cluster. The Pod range is fixed once assigned.
- B. Correct.
This is correct. GKE allows you to add an additional Pod range to an existing cluster using the
gcloud container clusters updatecommand, which avoids downtime and resolves IP exhaustion. - C. Incorrect.
This is incorrect because migrating the cluster to a new subnet would require recreating the cluster and would result in significant service disruption.
- D. Incorrect.
This is incorrect because recreating the cluster requires downtime and redeploying workloads, which is not the most efficient or minimal-disruption solution.