C3 - Part I - Loading tools
This page describes the Part I of the Challenge 3
Step 0 - Testing policy
Step 1 - JMeter
Download and install on your local machine the JMeter tool from http://jmeter.apache.org/.
Open two terminal windows side-by-side and, using SSH, log into each instance. Bring up a continuous display of the Apache access log by running the command sudo tail -F /var/log/apache2/access.log.
Using the AWS console, enable detailed (1-minute interval) monitoring of the two instances: Select an instance and click on the Monitoring tab. Click on Enable Detailed Monitoring.
Consult the JMeter documentation https://jmeter.apache.org/usermanual/build-web-test-plan.html and create a simple test plan. Specify the load balancer as the target for the HTTP requests. Run a test.
Observe which of the instances gets the load. Increase the load and re-run the test. Observe response times and time-outs. Repeat until you see unacceptable response times and/or time-outs.
Immediately after having created a high load for the site, re-run the nslookup command to resolve the DNS name of the load balancer into IP addresses to see if there are any changes.
Step 2 - Stress
Stress your instance and observe it with htop.
Observe the monitoring view on the AWS Console.
Step 3 - Analysis
When you resolve the DNS name of the load balancer into IP addresses while the load balancer is under high load what do you see? Explain.
Did this test really test the load balancing mechanism? What are the limitations of this simple test? What would be necessary to do realistic testing?
Last updated