Chapter 1 Introduction to Policies and Expressions 9

Order of Evaluation Based on Traffic Flow

As traffic flows through the NetScaler and is processed by various features, each
feature performs policy evaluation. Whenever a policy matches the traffic, the
NetScaler stores the action and continues processing until the data is about to
leave the NetScaler. At that point, the NetScaler typically applies all matching
actions. Integrated Caching, which only applies a final Cache or NoCache action,
is an exception.
Some policies affect the outcome of other policies. Following are examples:
• If a response is served from the Integrated Cache, some other NetScaler
features do not process the response or the request that initiated it.
• If the Content Filtering feature prevents a response from being served, no
subsequent features evaluate the response.
If the Application Firewall rejects an incoming request, no other features can
process it.
Advanced and Classic Expressions
One of the most fundamental components of a policy is its rule. A policy rule is a
logical expression that enables the policy to analyze traffic. Most of the policy's
functionality is derived from its expression.
An expression matches characteristics of traffic or other data with one or more
parameters and values. For example, an expression can enable the NetScaler to
accomplish the following:
• Determine whether a request contains a certificate.
• Determine the IP address of a client that sent a TCP request.
• Identify the data that an HTTP request contains (for example, a popular
spreadsheet or word processing application).
• Calculate the length of an HTTP request.

About Advanced Expressions

Any feature that uses advanced policies also uses advanced expressions. For
information on which features use advanced policies, see the table, “NetScaler
Feature, Policy Type, and Policy Usage,” on page 3.
Advanced expressions have a few other uses. In addition to configuring advanced
expressions in policy rules, you configure advanced expressions in the following
situations: