For a start we need to define the issue.
- How does it occur? -> Define the exact steps to take, systems used, configurations etc. (Scenario)
- What happens? -> Define deviation from expected behavior when the exact same steps are taken.
- What should happen? -> Define expected behavior.
- Who is affected? -> (Scope)
Now you have your baseline. All the steps that you take from now on should add information to Scenario or Scope.
Method 1: Basic and applicable for many cases.
- Follow the steps (scenario) to reproduce the behavior.
This will add information to the scope -> single local problem or multiple instances report the same behavior. - Change/Exclude 1 parameter - in example - use different laptop, system, account and observe if you get different results. If the results are the same - the parameter does not affect your problem. Repeat until you find a parameter that changes the results.
Repeat until you narrow it down so that you are able to fix it.
Method 2: Advanced and applicable for specific situations and systems.
- Capture logs - client, server, router, machine, your laptop.
- Analyze logs - That is the advanced part. You need to know how your system should work and be able to find the deviations that can lead to your problem.
- Identify differences and variations.
- Change ONE. Test.
- If issue is still present - repeat the steps again, until you have the desired results.