How can you test a web server network?

• Load testing: identifying maximum performance, stability, and scalability under high load.
• Functional testing: verifying correct request processing and response return by each server, load balancer functionality.
• Security testing: checking for vulnerabilities such as DDoS attacks, injections, data interception.
• Availability testing: testing failover when one server fails, overall system fault tolerance.
• Performance testing: measuring server response times, throughput.
• Configuration testing: verifying correct server, load balancer, network device settings.
• Monitoring: continuous control of server status, load, errors.

Tools:

• Nagios, Zabbix (monitoring)
• JMeter, LoadRunner (load testing)
• Nmap, Nessus (security testing)
• Ping, Traceroute (network diagnostics)

Example of a test scenario for availability check:

1. Send a request to the load balancer.
2. Ensure the request is redirected to one of the servers.
3. Simulate a failure of the active server (e.g., stop the process).
4. Send the next request to the load balancer.
5. Ensure the request is redirected to another available server.
6. Check the load balancer and server log files for errors.

Writing tests using Python and the requests library:

import requests
import time

# Load balancer availability test
def test_balancer_availability(url):
    try:
        response = requests.get(url)
        assert response.status_code == 200, f"Load balancer is unavailable or returned status {response.status_code}"
        print(f"Load balancer at {url} is available.")
    except requests.exceptions.RequestException as e:
        print(f"Error accessing load balancer: {e}")
        assert False, "Load balancer is unavailable"

# Load distribution test (checking that requests go to different servers)
def test_load_distribution(balancer_url, num_requests=10):
    responses = {}
    for _ in range(num_requests):
        try:
            response = requests.get(balancer_url)
            server_address = response.headers.get('X-Server-Address', 'Unknown') # Example header indicating server
            responses[server_address] = responses.get(server_address, 0) + 1
        except requests.exceptions.RequestException as e:
            print(f"Error requesting load balancer: {e}")
            pass # Ignore errors to continue testing

    print("Server request distribution:")
    for server, count in responses.items():
        print(f"- {server}: {count} requests")

    assert len(responses) > 1, "Requests are not distributed among multiple servers."
    print("Load distribution test passed.")

# Usage example
balancer_url = "http://your-load-balancer-address" # Replace with actual load balancer address

test_balancer_availability(balancer_url)
test_load_distribution(balancer_url)