2V0-71.23 Question 20
Select 3An organization deployed VMware Tanzu Kubernetes Grid (TKG) to manage Kubernetes clusters and is tasked with ensuring platform reliability and service-level agreements (SLAs) for critical workloads. As a Kubernetes platform administrator, which actions should you take to achieve this goal?
- A
Implement resource quotas and limits at the namespace level to control resource consumption.
- B
Deploy multiple replicas of critical application pods using Kubernetes Deployments or StatefulSets.
- C
Disable Kubernetes Role-Based Access Control (RBAC) to simplify user access management.
- D
Regularly monitor cluster health and resource usage through tools like VMware Tanzu Mission Control or Prometheus.
- E
Avoid setting up node affinity or anti-affinity rules to minimize complexity in scheduling.
Show answer and explanation
Correct answers: A, B, D
Explanation
To maintain platform reliability and meet SLAs for critical workloads, Kubernetes platform administrators should implement resource management (e.g., quotas and limits), ensure high availability through replication, and monitor the environment for potential issues. Disabling RBAC or neglecting scheduling rules like affinity/anti-affinity can compromise security and workload distribution, reducing platform reliability.
- A. Correct.
Implementing resource quotas and limits at the namespace level helps prevent resource overconsumption by individual workloads, ensuring the reliability of the platform for critical applications.
- B. Correct.
Deploying multiple replicas of critical application pods ensures high availability and fault tolerance, a key factor in meeting SLAs.
- C. Incorrect.
Disabling Kubernetes Role-Based Access Control (RBAC) is a security risk and does not align with best practices for platform administration.
- D. Correct.
Regularly monitoring cluster health and resource usage is essential for identifying potential issues early and maintaining platform reliability.
- E. Incorrect.
Avoiding node affinity or anti-affinity rules is not recommended because these rules are useful for distributing workloads effectively and avoiding resource contention.