Parallel execution in CI in Selenium Python - Build an Automation Script

Run two simple Selenium tests in parallel on CI

Preconditions (4)

Step 1: Create two simple Selenium test functions that open https://example.com and check the page title

Step 2: Configure pytest to run tests in parallel using pytest-xdist with 2 workers

Step 3: Push the test code and pytest configuration to the CI repository

Step 4: Trigger the CI pipeline to run the tests

Step 5: Observe that both tests run at the same time (in parallel) in the CI logs

Step 6: Verify that both tests pass successfully

✅ Expected Result: Both Selenium tests run simultaneously in the CI environment and pass, confirming parallel execution works.

Automation Requirements - pytest with Selenium WebDriver and pytest-xdist

Assertions Needed:

Verify page title is exactly 'Example Domain' in both tests

Verify both tests complete successfully in parallel

Best Practices:

Use explicit waits if needed (though example.com loads fast)

Use pytest fixtures for WebDriver setup and teardown

Avoid sharing WebDriver instances between tests

Use pytest-xdist -n 2 to run tests in parallel

Keep tests independent and stateless

Automated Solution

Selenium Python

import pytest
from selenium import webdriver
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options

@pytest.fixture
 def driver():
    options = Options()
    options.add_argument('--headless')
    options.add_argument('--no-sandbox')
    options.add_argument('--disable-dev-shm-usage')
    service = Service()
    driver = webdriver.Chrome(service=service, options=options)
    yield driver
    driver.quit()


def test_example_domain_title_1(driver):
    driver.get('https://example.com')
    assert driver.title == 'Example Domain'


def test_example_domain_title_2(driver):
    driver.get('https://example.com')
    assert driver.title == 'Example Domain'

# To run in parallel, use the command:
# pytest -n 2 test_parallel.py

The driver fixture creates a new Chrome WebDriver instance for each test, ensuring tests do not share browser sessions.

Both test_example_domain_title_1 and test_example_domain_title_2 open https://example.com and assert the page title is exactly 'Example Domain'.

Running pytest -n 2 test_parallel.py uses pytest-xdist to run the two tests in parallel with 2 workers.

This setup is ideal for CI environments where parallel execution speeds up test runs.

Headless Chrome options are used to run browsers without UI, suitable for CI servers.

Common Mistakes - 4 Pitfalls

Sharing a single WebDriver instance across multiple tests

Not using pytest-xdist or equivalent to enable parallelism

Hardcoding waits or using time.sleep instead of explicit waits

Running tests with visible browser windows in CI

Bonus Challenge

Now add data-driven testing with 3 different URLs and verify their titles in parallel

Show Hint

Practice

1. What is the main benefit of running Selenium tests in parallel during Continuous Integration (CI)?

easy

A. It reduces the total test execution time by running tests simultaneously.

B. It makes tests run slower but more accurately.

C. It ensures tests run only on one browser at a time.

D. It disables all tests except one to save resources.

Solution

1. Step 1: Understand parallel execution in CI

   Parallel execution means running multiple tests at the same time instead of one after another.

2. Step 2: Identify the benefit of parallel execution

   Running tests simultaneously saves time by reducing total test duration.

3. Final Answer:

   It reduces the total test execution time by running tests simultaneously. -> Option A

4. Quick Check:

   Parallel execution = faster testing [OK]

Hint: Parallel means running tests at the same time [OK]

Common Mistakes:

• Thinking parallel slows tests down
• Believing tests run only on one browser
• Confusing parallel with disabling tests

2. Which Python module is commonly used to run Selenium tests in parallel within a CI pipeline?

easy

A. ThreadPoolExecutor

B. unittest

C. multiprocessing

D. time

Solution

1. Step 1: Identify Python modules for parallel execution

   Python offers several modules, but ThreadPoolExecutor from concurrent.futures is popular for parallel threads.

2. Step 2: Match module to Selenium parallel testing

   ThreadPoolExecutor allows running multiple Selenium tests in parallel threads efficiently.

3. Final Answer:

   ThreadPoolExecutor -> Option A

4. Quick Check:

   ThreadPoolExecutor = parallel threads [OK]

Hint: ThreadPoolExecutor is for parallel threads in Python [OK]

Common Mistakes:

• Choosing multiprocessing which uses processes, not threads
• Selecting unittest which is a test framework, not parallel executor
• Picking time which is for delays, not parallelism

3. Given the code snippet below, what will be the output?

from concurrent.futures import ThreadPoolExecutor
import time

def test_func(name):
    time.sleep(1)
    return f"Test {name} done"

with ThreadPoolExecutor(max_workers=3) as executor:
    results = list(executor.map(test_func, ['A', 'B', 'C', 'D']))
print(results)

medium

A. Error due to max_workers less than tasks

B. ['Test A done', 'Test B done', 'Test C done']

C. ['Test A done', 'Test B done', 'Test C done', 'Test D done']

D. ['Test D done']

Solution

1. Step 1: Understand ThreadPoolExecutor with max_workers=3

   It runs up to 3 threads at once, but can handle more tasks by queuing them.

2. Step 2: Analyze executor.map behavior

   executor.map runs test_func on all 4 inputs and returns results in order after all finish.

3. Final Answer:

   ['Test A done', 'Test B done', 'Test C done', 'Test D done'] -> Option C

4. Quick Check:

   All tasks run, results list includes all [OK]

Hint: ThreadPoolExecutor queues extra tasks beyond max_workers [OK]

Common Mistakes:

• Assuming only 3 results because max_workers=3
• Expecting error due to more tasks than workers
• Thinking results order changes

4. You wrote this code to run Selenium tests in parallel but get an error: RuntimeError: cannot schedule new futures after shutdown. What is the likely cause?

from concurrent.futures import ThreadPoolExecutor
executor = ThreadPoolExecutor(max_workers=2)
executor.shutdown()
executor.submit(test_function)

medium

A. ThreadPoolExecutor requires a context manager to work.

B. max_workers=2 is too low for submitting tasks.

C. test_function is not defined before submit call.

D. Calling submit after shutdown closes the executor, so no new tasks allowed.

Solution

1. Step 1: Understand executor.shutdown() effect

   shutdown() stops the executor from accepting new tasks and cleans up threads.

2. Step 2: Analyze submit call after shutdown

   Calling submit after shutdown causes RuntimeError because executor is closed.

3. Final Answer:

   Calling submit after shutdown closes the executor, so no new tasks allowed. -> Option D

4. Quick Check:

   submit after shutdown = error [OK]

Hint: Don't submit tasks after executor.shutdown() [OK]

Common Mistakes:

• Thinking max_workers causes this error
• Assuming test_function undefined causes this error
• Believing context manager is required

5. You want to run Selenium tests in parallel on Chrome and Firefox browsers in your CI pipeline using Python. Which approach correctly sets up parallel execution for both browsers?

hard

A. Run tests sequentially by switching browser drivers inside a single test function.

B. Use ThreadPoolExecutor to run a test function twice, each time initializing a different browser driver inside the function.

C. Use a single WebDriver instance shared by all tests to save memory.

D. Run tests in parallel but initialize browsers outside the test functions to avoid overhead.

Solution

1. Step 1: Understand parallel testing with multiple browsers

   Each test must create its own browser driver instance to avoid conflicts.

2. Step 2: Use ThreadPoolExecutor to run tests in parallel

   Running test functions in parallel, each initializing its own browser, allows simultaneous testing on Chrome and Firefox.

3. Final Answer:

   Use ThreadPoolExecutor to run a test function twice, each time initializing a different browser driver inside the function. -> Option B

4. Quick Check:

   Separate drivers per thread = correct parallel setup [OK]

Hint: Initialize browser inside each parallel test function [OK]

Common Mistakes:

• Sharing one WebDriver instance across threads
• Switching browsers sequentially defeats parallelism
• Initializing browsers outside test functions causes conflicts