Desired replicas vs actual replicas in Kubernetes - Practice Questions

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding Desired vs Actual Replicas in Kubernetes

In Kubernetes, what does the desired replicas count represent compared to the actual replicas count?

A Desired replicas is the number of pods the user wants running; actual replicas is the number currently running.

B Desired replicas is the number of pods currently running; actual replicas is the number the user wants running.

C Desired replicas and actual replicas always have the same value and mean the same thing.

D Desired replicas is the number of nodes in the cluster; actual replicas is the number of pods running.

💡 Hint

Think about what you tell Kubernetes to do versus what it has done so far.

The desired replicas is the number of pod instances you ask Kubernetes to maintain. The actual replicas is how many pods are currently running. Kubernetes tries to match actual to desired.

💻 Command Output

intermediate

Interpreting kubectl get deployment Output

Given the output below from kubectl get deployment myapp, what is the actual number of pods running?

NAME    READY   UP-TO-DATE   AVAILABLE   AGE
myapp   3/5     5            3           10m

A 3 pods are desired but 5 are running.

B 5 pods are running and ready.

C 3 pods are running and ready.

D 5 pods are available but only 3 are up-to-date.

💡 Hint

The READY column shows how many pods are ready out of desired.

The READY column shows 3/5, meaning 3 pods are ready and running out of 5 desired replicas.

🔀 Workflow

advanced

Troubleshooting Replica Mismatch in Deployment

You notice your Kubernetes deployment's desired replicas is 4 but actual replicas is stuck at 2. Which step should you take first to diagnose the issue?

A Check pod status with <code>kubectl get pods</code> to see if pods are failing or pending.

B Delete the deployment and recreate it immediately.

C Scale the deployment to 10 replicas to force new pods.

D Restart the Kubernetes cluster to reset pod states.

💡 Hint

Look at the pods themselves before making big changes.

Checking pod status helps identify if pods are crashing or stuck, which explains why actual replicas don't match desired.

❓ Configuration

advanced

Correcting Replica Count in Deployment YAML

Which snippet correctly sets the desired replicas to 3 in a Kubernetes deployment YAML?

A

metadata:
  replicas: 3
spec:
  template:
    metadata:
      labels:
        app: myapp

B

spec:
  replicas: 3
  template:
    metadata:
      labels:
        app: myapp

C

spec:
  template:
    replicas: 3
    metadata:
      labels:
        app: myapp

D

replicas: 3
spec:
  template:
    metadata:
      labels:
        app: myapp

💡 Hint

Replicas is a field under spec, not metadata or template.

The replicas field must be directly under spec to set desired pod count. Other placements are invalid.

❓ Troubleshoot

expert

Diagnosing Replica Count Not Scaling Up

You updated your deployment to increase replicas from 2 to 5, but kubectl get deployment still shows 2 actual replicas after 10 minutes. What is the most likely cause?

A The container image tag is incorrect causing pods to crash immediately.

B The deployment YAML file was not saved before applying changes.

C The Kubernetes API server is down and not processing requests.

D Pods are stuck in Pending state due to insufficient cluster resources.

💡 Hint

Think about what prevents new pods from starting even if desired replicas increased.

If cluster resources like CPU or memory are insufficient, new pods remain Pending and actual replicas stay low.