Methodology vs. Approaches/styles

• A methodology describes a prescribed path taken to systematically solve problems
• An approach or technique is a way to gather information by addressing heuristically targeting specific network elements first

Troubleshooting Methodologies

Cisco Official Troubleshooting Methodology

The Official Cisco Troubleshooting Methodology^[1] is an 8-step process that loops back and forth around itself.

Always (always, at any step, for any reason) DOCUMENT any and all changes that have been made.

1. Define the problem in terms of symptoms and potential causes
2. Gather facts
   3. Ask affected users, and collect information from NMS, PCAPs, output from Diagnostic Commands, etc.
3. Analyze information and eliminate potential causes
   1. The elimination of potential causes is critical to create an efficient action plan
4. Create an action plan which changes only one variable at a time
   1. This plan should target potential causes from most likely to least likely
5. Implement Action Plan
   1. Document all changes made
6. Gather results of each change
7. Analyze the results to determine if issue is resolved
   1. If resolved, document changes and process is complete.
8. If unresolved, return to step 4

In an effort to make this more visually comprehensible, I've drafted this flow-chart.

Source: Original

Flackbox Troubleshooting Methodology

Neil Anderson with FlackBox proposes a similar but different strategy. I've decided to include it here because it may be easier for you to memorize, and is basically the same steps, but expanded.

It also makes a handy acronym, DGAE-PTSD

1. Define the problem
2. Gather information
3. Analyze information
4. Eliminate potential causes
5. Propose hypothesis
6. Test Hypothesis
7. Solve Problem and Document Solution

Source: FlackBox

Troubleshooting Approaches/Techniques

There are 4 key troubleshooting approaches/techniques you should use

1. Work around the OSI model
   1. Top down
      1. Application>Physical
   2. Bottom up
      1. Physical>Application
   3. Divide and Conquer
      1. Start in the middle
2. Compare device configurations
   1. Comparing configurations between backups and devices, running vs. startup, etc.
   2. Can identify ad hoc or undocumented changes that impacted performance
3. Trace the path
   1. Trace the path from source to destination
   2. Helpful in identifying Layer 1 issues
4. Check for broken components
   1. Is the equipment is old?
   2. Was there a recent event (power surge, HVAC issue, liquid damage, etc) that might have caused an issue?
   3. Did a fiber cable get pinched or an Ethernet cable get cut?

Metadata

OSI or TCP/IP Layer

CCNA Exam Topic

Contributors

Sources

Troubleshooting Overview Cisco Systems
The Cisco Troubleshooting Methodology - FlackBox