kubectl apply vs create in Kubernetes - Visual Side-by-Side Comparison

Process Flow - kubectl apply vs create

Write YAML file

↓

kubectl create

→Create resource if not exists

↓

Error if exists

↓

kubectl apply

→Create resource if not exists

↓

Update resource if exists

This flow shows how 'kubectl create' only makes new resources and errors if they exist, while 'kubectl apply' creates or updates resources based on YAML files.

Execution Sample

Kubernetes

kubectl create -f pod.yaml
kubectl apply -f pod.yaml

Creates a pod resource with 'create', then updates or creates it with 'apply'.

Process Table

Step	Command	Resource Exists?	Action Taken	Output/Result
1	kubectl create -f pod.yaml	No	Create resource	pod/my-pod created
2	kubectl create -f pod.yaml	Yes	Error	Error from server (AlreadyExists): pods "my-pod" already exists
3	kubectl apply -f pod.yaml	No	Create resource	pod/my-pod created
4	kubectl apply -f pod.yaml	Yes	Update resource	pod/my-pod configured

💡 kubectl create errors if resource exists; kubectl apply creates or updates resource.

Status Tracker

Resource State	Start	After Step 1	After Step 2	After Step 3	After Step 4
pod/my-pod	Does not exist	Created	No change (error)	Created	Updated

Key Moments - 2 Insights

Why does 'kubectl create' fail on the second attempt?

'kubectl create' fails because the resource already exists, as shown in step 2 of the execution table where it returns an error.

How does 'kubectl apply' handle existing resources differently?

'kubectl apply' updates the existing resource instead of erroring, as seen in step 4 where it says 'pod/my-pod configured'.

Visual Quiz - 3 Questions

Look at the execution table, what happens when you run 'kubectl create' on an existing resource?

A It updates the resource

B It deletes the resource

C It returns an error

D It ignores the command

💡 Hint

Check step 2 in the execution table where 'kubectl create' is run on an existing resource.

At which step does 'kubectl apply' update an existing resource?

A Step 4

B Step 2

C Step 1

D Step 3

💡 Hint

Look at the 'Action Taken' column for 'kubectl apply' in the execution table.

If the resource does not exist, what do both 'kubectl create' and 'kubectl apply' do?

A Both return an error

B Both create the resource

C 'create' errors, 'apply' creates

D 'apply' errors, 'create' creates

💡 Hint

Check steps 1 and 3 in the execution table for commands run when resource does not exist.

Concept Snapshot

kubectl create -f file.yaml
  - Creates resource only if it does not exist
  - Errors if resource exists

kubectl apply -f file.yaml
  - Creates resource if missing
  - Updates resource if exists

Use 'apply' for managing config changes safely.

Full Transcript

This visual shows the difference between 'kubectl create' and 'kubectl apply'. 'kubectl create' makes a resource only if it does not exist and errors if it does. 'kubectl apply' creates the resource if missing or updates it if it exists. The execution table traces commands on a pod resource, showing creation, error on duplicate create, and update with apply. Variables track the pod's existence state. Key moments clarify why create errors on duplicates and how apply updates. The quiz tests understanding of these behaviors. Remember, use 'kubectl apply' to safely manage resource changes.