Why Resource quotas per namespace in Kubernetes? - Purpose & Use Cases

The Big Idea

What if one team's app crashes the whole cluster because it used too many resources? Resource quotas stop that from happening.

The Scenario

Imagine you manage a shared playground where many kids want to play with limited toys. Without rules, some kids grab all the toys, leaving none for others.

The Problem

Manually tracking who uses what resources in a Kubernetes cluster is like watching every kid all the time. It's slow, tiring, and mistakes happen--some teams use too much, others get none.

The Solution

Resource quotas per namespace set clear limits on how much each team can use. This way, everyone gets a fair share automatically, and the system keeps things balanced without constant watching.

Before vs After

✗ Before

No limits set; teams deploy freely and can consume all resources.

✓ After

apiVersion: v1
kind: ResourceQuota
metadata:
  name: team-quota
  namespace: team-namespace
spec:
  hard:
    pods: '10'
    requests.cpu: '4'
    requests.memory: 8Gi
    limits.cpu: '8'
    limits.memory: 16Gi

What It Enables

It enables fair, automatic sharing of cluster resources so no team can accidentally block others.

Real Life Example

A company runs multiple projects in one Kubernetes cluster. Using resource quotas per namespace, each project gets a fixed amount of CPU and memory, preventing one project from crashing the whole system.

Key Takeaways

Manual resource tracking is slow and error-prone.

Resource quotas set clear, automatic limits per namespace.

This keeps cluster resources fairly shared and stable.

Practice

1. What is the main purpose of a ResourceQuota in a Kubernetes namespace?

easy

A. To limit the amount of resources like CPU and memory used in that namespace

B. To increase the CPU speed of all pods in the namespace

C. To delete unused pods automatically

D. To backup data stored in the namespace

Solution

1. Step 1: Understand ResourceQuota purpose

   A ResourceQuota sets limits on resource usage like CPU, memory, pods, and storage within a namespace.

2. Step 2: Compare options with purpose

   Only To limit the amount of resources like CPU and memory used in that namespace correctly describes limiting resource usage. Other options describe unrelated actions.

3. Final Answer:

   To limit the amount of resources like CPU and memory used in that namespace -> Option A

4. Quick Check:

   ResourceQuota limits resources = D [OK]

Hint: ResourceQuota controls resource limits per namespace [OK]

Common Mistakes:

• Thinking ResourceQuota speeds up pods
• Confusing quotas with backups
• Assuming quotas delete pods automatically

2. Which of the following is the correct YAML snippet to define a CPU limit of 2 cores in a ResourceQuota?

easy

A. hard:
limits.cpu: "2"

B. resources:
limits:
cpu: "2"

C. hard:
cpu: "2"

D. hard:
limits.cpu: 2

Solution

1. Step 1: Recall ResourceQuota syntax for limits

   ResourceQuota uses the hard field with resource names like limits.cpu as keys and string values.

2. Step 2: Check each option's syntax

   hard: limits.cpu: "2" correctly uses hard: with limits.cpu: "2". resources: limits: cpu: "2" uses wrong keys, C misses limits., D uses number instead of string.

3. Final Answer:

   hard: limits.cpu: "2" -> Option A

4. Quick Check:

   ResourceQuota hard limits use string values with resource keys [OK]

Hint: ResourceQuota limits use 'hard' with string values [OK]

Common Mistakes:

• Using numeric values instead of strings
• Omitting 'limits.' prefix in resource keys
• Placing limits under wrong YAML keys

3. Given this ResourceQuota YAML applied to namespace dev:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: dev-quota
  namespace: dev
spec:
  hard:
    pods: "3"
    requests.cpu: "1"
    requests.memory: "1Gi"

What happens if you try to create a 4th pod in dev namespace?

medium

A. The pod will be created normally without any error

B. The pod will be created but with limited CPU

C. The pod creation will fail due to exceeding the pod count quota

D. The pod will be created but memory will be limited to 1Gi

Solution

1. Step 1: Understand pod count limit in ResourceQuota

   The quota limits pods to 3 in the dev namespace.

2. Step 2: Effect of creating 4th pod

   Creating a 4th pod exceeds the pod count quota, so Kubernetes rejects the creation with an error.

3. Final Answer:

   The pod creation will fail due to exceeding the pod count quota -> Option C

4. Quick Check:

   Pod count quota exceeded = creation fails [OK]

Hint: Pod count quota blocks pods beyond limit [OK]

Common Mistakes:

• Assuming pod is created with limited resources
• Thinking quota only warns but allows creation
• Confusing CPU/memory limits with pod count

4. You applied this ResourceQuota YAML but it fails to create:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: test-quota
  namespace: test
spec:
  hard:
    cpu: "2"
    memory: "4Gi"

What is the likely error?

medium

A. Missing 'apiVersion' field

B. Namespace 'test' does not exist

C. Values should be numbers, not strings

D. Resource keys 'cpu' and 'memory' are invalid; should use 'limits.cpu' and 'limits.memory'

Solution

1. Step 1: Check resource keys in ResourceQuota

   ResourceQuota requires full resource names like limits.cpu or requests.memory, not just cpu or memory.

2. Step 2: Validate YAML fields

   API version is present, namespace existence is unrelated to syntax error, and values must be strings (which they are).

3. Final Answer:

   Resource keys 'cpu' and 'memory' are invalid; should use 'limits.cpu' and 'limits.memory' -> Option D

4. Quick Check:

   Use full resource keys like limits.cpu [OK]

Hint: Use full resource keys like limits.cpu in quotas [OK]

Common Mistakes:

• Using short keys like 'cpu' instead of 'limits.cpu'
• Assuming missing namespace causes syntax error
• Using numeric values instead of strings

5. You want to limit a namespace to 5 pods and 2Gi memory requests, but allow unlimited CPU. Which ResourceQuota YAML snippet achieves this?

hard

A. spec:
hard:
pods: "5"
requests.memory: "2Gi"
requests.cpu: "0"

B. spec:
hard:
pods: "5"
requests.memory: "2Gi"

C. spec:
hard:
pods: "5"
requests.memory: "2Gi"
limits.cpu: "unlimited"

D. spec:
hard:
pods: "5"
requests.memory: "2Gi"
requests.cpu: "-1"

Solution

1. Step 1: Understand ResourceQuota limits

   To limit pods and memory requests, specify those keys under hard. Omitting CPU means no limit on CPU.

2. Step 2: Evaluate options for CPU limits

   spec: hard: pods: "5" requests.memory: "2Gi" omits CPU limits, allowing unlimited CPU. Options A, C, D try to set CPU limits with invalid or restrictive values.

3. Final Answer:

   spec: hard: pods: "5" requests.memory: "2Gi" -> Option B

4. Quick Check:

   Omit CPU limit to allow unlimited CPU [OK]

Hint: Omit resource in quota to allow unlimited usage [OK]

Common Mistakes:

• Setting CPU limit to 0 or negative disables pods
• Using 'unlimited' as a value (not valid)
• Trying to limit CPU when unlimited is desired