Troubleshooting guide

2-5
Cisco Broadband Local Integrated Services Solution Troubleshooting Guide
OL-5169-01
Chapter 2 Troubleshooting Overview
Troubleshooting Strategy
Detailed Troubleshooting Methodology
The goal of this section is to establish a methodical mindset—an ordered pattern of thought to use when
troubleshooting. The model described in this section takes a multistep approach to problem-solving. In
the following subsections, we will examine each of these steps in detail to see how it can be used in a
troubleshooting example.
Step 1: Define the Problem
A systematic approach to troubleshooting consists of a sequence of steps. The first grouping of these
steps is to make sure you have a clear, sufficient definition of the problem. When analyzing a network
problem, make a clear problem statement by defining the problem in terms of a set of symptoms and
associated causes. To do this, identify the general symptoms and ascertain what possible causes could
result in these symptoms.
At this point in the methodology, you define the problem by identifying associated general symptoms
and identifying possible causes. Try to form opinions of possible causes and document them. Many
answers might arise, but concentrate on those that could be considered major contributors to the
problem. Subsequent steps in the methodology allow you to ask questions (that is, gather facts).
Form problem statements with reference to the baselines you have established for your network. You
should know what your network indicators look like when the network is performing as you expect. Also,
you must have knowledge of any network changes since the last evidence of baseline performance.
In this group of steps, you should gather the initial diagnosis made by the end users. What the end user
reports is important; however, the full definition of the problem may have a broader basis. If possible,
proceed from your own knowledge about your network and try to see the problem for yourself.
As part of the systematic approach to troubleshooting, many support teams have developed a set of
primary questions and processes to use when getting problem report information from end users. Among
the primary questions are "How often has this problem happened?" and "When did it start?" and "Can
you readily reproduce the problem condition, and if so, how?"
Step 2: Gather the Facts
The second step in troubleshooting is to gather the facts you need to help isolate possible causes.
Ask questions of affected users, network administrators, managers, and any other key people involved
with the network. Try to ascertain whether anyone is aware of anything that has changed. Thoroughly
document all information received.
Depending on the nature of the reported symptoms, collect facts from sources such as network
management systems, protocol analyzer traces, output from router diagnostic commands such as debug
commands and show commands, or software release notes. It might be necessary to collect this
information at discrete times or over extended time periods, such as an overnight data capture.
It is always a good idea to document and keep on record copies of the configurations of switches, routers,
servers, and any other configurable network devices to be able to compare configurations and determine
whether anything has changed. You need to gather facts in an attempt to focus on the possible causes.
Step 3: Consider the Possibilities
Using the data you gathered and your knowledge of the systems and devices in your network, you can
set the boundaries that help you begin to isolate the problem cause(s). By setting boundaries, you focus
on only those portions of the solution that are relevant to the specific problem or failure mode.