API authentication (OAuth, JWT, API keys) in HLD - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding OAuth 2.0 Flow

Which step in the OAuth 2.0 Authorization Code flow is responsible for exchanging the authorization code for an access token?

A The client redirects the user to the authorization server's login page.

B The resource owner grants permission to the client application.

C The client sends the authorization code to the authorization server's token endpoint.

D The client uses the access token to access protected resources.

💡 Hint

Think about which step involves the client securely obtaining the token after user approval.

In OAuth 2.0 Authorization Code flow, after the user grants permission, the client receives an authorization code. The client then sends this code to the authorization server's token endpoint to exchange it for an access token.

❓ Architecture

intermediate

Choosing API Key Storage Location

Where is the most secure place to store API keys in a client-server architecture to prevent unauthorized access?

A Store API keys in environment variables on the server side.

B Store API keys in the client-side JavaScript code for easy access.

C Embed API keys directly in HTML meta tags.

D Store API keys in local storage of the user's browser.

💡 Hint

Consider where secrets should be kept hidden from users.

API keys should be stored securely on the server side, such as in environment variables, to prevent exposure to clients or users. Client-side storage exposes keys to anyone who can inspect the code or browser storage.

❓ scaling

advanced

Scaling JWT Validation in a Distributed System

In a distributed system with multiple API servers validating JWT tokens, what is the best approach to ensure consistent and scalable token validation?

A API servers store the JWT secret key and generate tokens themselves.

B Each API server fetches the public key from the authorization server on every request.

C API servers validate tokens by querying the authorization server synchronously for each request.

D API servers cache the public key locally and refresh it periodically.

💡 Hint

Think about reducing latency and load on the authorization server.

Caching the public key locally on API servers and refreshing it periodically reduces latency and avoids overloading the authorization server. Fetching on every request or synchronous validation causes performance bottlenecks.

❓ tradeoff

advanced

Tradeoffs Between OAuth and API Keys

Which statement best describes a key tradeoff when choosing between OAuth 2.0 and API keys for API authentication?

A OAuth provides fine-grained access control but is more complex to implement than API keys.

B OAuth tokens never expire, making them simpler to manage than API keys.

C API keys support delegated access and user consent flows better than OAuth.

D API keys are more secure than OAuth because they do not require token exchange.

💡 Hint

Consider complexity and control features of each method.

OAuth 2.0 supports delegated access with user consent and fine-grained permissions but requires more setup and complexity. API keys are simpler but less flexible and secure.

❓ estimation

expert

Estimating Token Storage Requirements for JWTs

You expect 1 million active users, each making 10 API requests per minute. Each JWT token is 1 KB in size and stored in a cache for 15 minutes. Approximately how much memory (in GB) is needed to store all active tokens in the cache at peak usage?

A Approximately 10 GB

B Approximately 150 GB

C Approximately 250 GB

D Approximately 100 GB

💡 Hint

Calculate how many unique tokens are active in the cache at once and multiply by token size.

Each user has one active token cached for 15 minutes. With 1 million users, total tokens cached = 1 million. Each token is 1 KB, so total memory = 1 million KB = ~1 GB. However, if tokens are refreshed every request (10 per minute), and cache holds tokens for 15 minutes, total tokens = 1 million users * 10 requests/min * 15 min = 150 million tokens, which is 150 million KB = ~150 GB. The question implies caching tokens per request, so 150 GB is correct.