How to Mock Fault Injection Percentage in Python

Simulate a service with a 30% failure rate using random.random to test error handling and retries.

Easy Python 3.9+ Aug 9, 2026 Reliability & rate limiting 14 views 0 copies

fault-injection random testing reliability error-handling

Python code

21 lines

Python 3.9+

import random

class Service:
    def call(self):
        if random.random() < 0.3:  # 30% failure rate
            raise ConnectionError("Simulated network fault")
        return "ok"

def main():
    svc = Service()
    random.seed(42)  # deterministic for demonstration
    results = []
    for _ in range(10):
        try:
            results.append(svc.call())
        except ConnectionError:
            results.append("error")
    print(results)

if __name__ == "__main__":
    main()

Output

stdout

['error', 'ok', 'ok', 'ok', 'ok', 'ok', 'error', 'ok', 'ok', 'ok']

How it works

The random.random() call returns a float between 0 and 1, and comparing it to 0.3 gives roughly a 30% chance of raising ConnectionError. Seeding with random.seed(42) makes the failure pattern reproducible for testing. Each iteration calls svc.call() inside a try/except to capture errors as strings, simulating how a caller would handle partial failures. This pattern is useful for testing retry logic and circuit breakers without real network dependencies.

Common mistakes

Forgetting to seed random for reproducible tests, making failures nondeterministic
Using `random.randint` instead of `random.random` for probability-based faults
Not wrapping calls in try/except, causing the script to crash on the first simulated failure
Hardcoding the percentage in multiple places instead of as a configurable parameter

Variations

Use `random.random() < failure_rate` with failure_rate passed as a function argument for configurable fault injection
Use a decorator or wrapper that injects faults based on a counter for a percentage of total calls

Real-world use cases

Testing retry and backoff logic in a microservice client to ensure it recovers gracefully from transient network errors.
Validating circuit breaker behavior by injecting faults at a controlled rate in a load test environment.
Verifying monitoring and alerting systems trigger correctly when a service experiences a predefined failure percentage.

How to Mock Fault Injection Percentage in Python

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Reliability & rate limiting

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Reliability & rate limiting

Keep learning

Tutorials

Quizzes