Environment variable management in FastAPI - Performance & Optimization

Performance: Environment variable management

MEDIUM IMPACT

This affects the startup time and configuration loading speed of the FastAPI application.

Loading environment variables for configuration in a FastAPI app

FastAPI

import os
from dotenv import load_dotenv

load_dotenv()  # called once at startup

secret = os.getenv('SECRET_KEY')

def read_root():
    return {"secret": secret}

Loading environment variables once at startup avoids repeated blocking calls during requests, improving response speed.

📈 Performance GainSingle blocking load at startup, no blocking during requests, improves LCP and INP.

Loading environment variables for configuration in a FastAPI app

FastAPI

import os
from dotenv import load_dotenv

def read_root():
    load_dotenv()  # called inside a route handler
    secret = os.getenv('SECRET_KEY')
    return {"secret": secret}

Calling load_dotenv inside a route handler causes environment variables to be loaded on every request, blocking response and increasing latency.

📉 Performance CostBlocks rendering for every request, increasing INP and LCP significantly.

Performance Comparison

Pattern	DOM Operations	Reflows	Paint Cost	Verdict
Load env vars inside route	N/A	N/A	Blocks response start, delays paint	[X] Bad
Load env vars once at startup	N/A	N/A	Fast response start, no blocking	[OK] Good
Repeated os.getenv calls in handler	N/A	N/A	Minor CPU overhead, slight delay	[!] OK
Cache env vars in constants	N/A	N/A	Minimal overhead, fast response	[OK] Good

Rendering Pipeline

Environment variable loading happens before the rendering pipeline starts. If done inefficiently during request handling, it delays the start of rendering and response.

→App Initialization

→Request Handling

⚠️ BottleneckBlocking environment variable loading during request handling delays response start.

Core Web Vital Affected

LCP

This affects the startup time and configuration loading speed of the FastAPI application.

Optimization Tips

1Load environment variables once at app startup, not per request.

2Cache environment variable values in constants for fast access.

3Avoid blocking calls during request handling to improve LCP and INP.

Performance Quiz - 3 Questions

What is the best practice for loading environment variables in a FastAPI app to optimize performance?

A Load them inside every route handler

B Load them once at app startup and cache values

C Load them only when a user requests a specific route

D Reload them from disk on every request

DevTools: Performance

How to check: Record a profile while making requests to your FastAPI app. Look for blocking calls or delays before the first byte is sent.

What to look for: Long blocking tasks or delays in the startup or request handling phase indicate inefficient environment variable loading.

Practice

1. What is the main reason to use environment variables in a FastAPI project?

easy

A. To keep sensitive information like API keys out of the code

B. To make the code run faster

C. To store user data permanently

D. To create new API endpoints automatically

Solution

1. Step 1: Understand environment variables purpose

   Environment variables are used to keep sensitive data like passwords or API keys separate from the code to improve security.

2. Step 2: Identify the correct reason in FastAPI context

   FastAPI uses environment variables to avoid hardcoding secrets, which helps keep them safe and flexible.

3. Final Answer:

   To keep sensitive information like API keys out of the code -> Option A

4. Quick Check:

   Environment variables = keep secrets safe [OK]

Hint: Environment variables store secrets, not speed or data [OK]

Common Mistakes:

• Thinking environment variables speed up code
• Confusing environment variables with database storage
• Assuming environment variables create API routes

2. Which of the following is the correct way to get an environment variable named API_KEY using Python's os module in FastAPI?

easy

A. api_key = os.get('API_KEY')

B. api_key = os.getenv('API_KEY')

C. api_key = os.environ.get('API_KEY')()

D. api_key = os.getenv(API_KEY)

Solution

1. Step 1: Recall correct os module function

   The correct function to get environment variables is os.getenv('VAR_NAME').

2. Step 2: Check syntax correctness

   api_key = os.getenv('API_KEY') uses os.getenv('API_KEY') correctly with quotes around the variable name and no extra parentheses.

3. Final Answer:

   api_key = os.getenv('API_KEY') -> Option B

4. Quick Check:

   Use os.getenv('VAR') to get environment variables [OK]

Hint: Use os.getenv('VAR_NAME') with quotes [OK]

Common Mistakes:

• Using os.get instead of os.getenv
• Calling os.environ.get with extra parentheses
• Not quoting the environment variable name

3. Given this FastAPI code snippet using Pydantic's BaseSettings:

from pydantic import BaseSettings

class Settings(BaseSettings):
    app_name: str
    debug_mode: bool = False

settings = Settings()

If the environment variable APP_NAME is set to "MyApp" and DEBUG_MODE is not set, what will settings.app_name and settings.debug_mode be?

medium

A. None and None

B. "MyApp" and True

C. None and False

D. "MyApp" and False

Solution

1. Step 1: Understand BaseSettings behavior

   Pydantic BaseSettings reads environment variables matching attribute names in uppercase by default.

2. Step 2: Check provided environment variables and defaults

   APP_NAME is set to "MyApp", so settings.app_name is "MyApp". DEBUG_MODE is not set, so it uses the default False.

3. Final Answer:

   "MyApp" and False -> Option D

4. Quick Check:

   Env var set = value, missing uses default [OK]

Hint: Unset vars use default values in BaseSettings [OK]

Common Mistakes:

• Assuming missing env vars become True
• Thinking unset vars become None instead of default
• Confusing attribute names with env var names

4. What is wrong with this FastAPI environment variable usage code?

import os

API_KEY = os.getenv('API_KEY')
print(f"API key is {API_KEY.upper()}")

Assuming API_KEY is not set in the environment.

medium

A. It will raise an AttributeError because API_KEY is None

B. It will print 'API key is ' with empty string

C. It will raise a SyntaxError

D. It will print the string 'None'

Solution

1. Step 1: Check os.getenv behavior when var missing

   os.getenv returns None if the environment variable is not set and no default is given.

2. Step 2: Understand what happens calling upper() on None

   Calling None.upper() raises an AttributeError because NoneType has no method upper().

3. Final Answer:

   It will raise an AttributeError because API_KEY is None -> Option A

4. Quick Check:

   None.upper() causes AttributeError [OK]

Hint: Check for None before calling string methods [OK]

Common Mistakes:

• Assuming os.getenv returns empty string if missing
• Expecting no error when calling upper() on None
• Confusing AttributeError with SyntaxError

5. You want to manage environment variables in FastAPI using Pydantic's BaseSettings and ensure a default port of 8000 if PORT is not set. Which code snippet correctly achieves this?

hard

A. class Settings(BaseSettings):
port: int

settings = Settings(port=8000)

B. class Settings(BaseSettings):
port: int

settings = Settings() # default 8000 if missing

C. class Settings(BaseSettings):
port: int = 8000

settings = Settings()

D. class Settings(BaseSettings):
port: str = '8000'

settings = Settings()

Solution

1. Step 1: Define default value in BaseSettings attribute

   Setting port: int = 8000 provides a default if the environment variable PORT is missing.

2. Step 2: Instantiate Settings without arguments

   Calling Settings() reads environment variables or uses defaults if missing, so port will be 8000 if PORT is unset.

3. Final Answer:

   class Settings(BaseSettings): port: int = 8000 settings = Settings() -> Option C

4. Quick Check:

   Default values in BaseSettings = default used if env var missing [OK]

Hint: Set default in class attribute for fallback [OK]

Common Mistakes:

• Not providing default and expecting fallback
• Passing default in constructor instead of attribute
• Using string type for port instead of int