Juniper NETWORK AND SECURITY MANAGER 2010.4 - ADMININISTRATION GUIDE REV1, Руководство администратора

Страница: 526 / 1026

protocol is a specification that indicates how communication between two entities
(applications, servers, Ethernet cards, etc.) occurs.
When attacking a system, attackers use the protocol of a supported service to
communicate their malicious activity to the server. However, attackers can only use
protocols that are supported by the system they are attacking. You can add a category
group to the Attacks column in your rule; however, you need to select only the categories
that are used by the address objects you are protecting with the rule.
For example, if you rely extensively on FTP and HTTP for file transfers to and from your
Web servers, choose the FTP and HTTP category groups to carefully monitor all traffic
that uses these services.
NOTE: As of Release 2007.3, a few of the entries in the IDP attack group
table, starting with the Response category, are removed to enhance the
performance of IDP devices. See the latest NSM Release Notes for information
on the Response category removed from the IDP attack group table.
Adding IDP Attack Objects by Operating System
The Operating System group includes attack objects for several predefined operating
systems to help you choose the attack objects that are the most dangerous to specific
components on your network. You can choose BSD, Linux, Solaris, or Windows.
Adding IDP Attack Objects by Severity
The Severity group includes five attack object groups organized by severity level. You
can select one or more groups to include in your rule. To protect critical address objects
or “popular” attacker targets, such as your mail server, use multiple severity levels to
ensure maximum protection.
We recommend using the following actions and notification settings listed in Table 44
on page 476 when using severity-based dynamic attack groups in a rule:
Table 44: Severity Levels, Recommended Actions and Notifications
Notification Recommended Action Cause Severity
Logging
Alert
Drop Packet Attacks attempt to evade an IDS, crash a machine, or gain
system-level privileges. Critical
Logging
Alert
Drop Packet
Drop Connection
Attacks attempt to crash a service, perform a denial-of-service,
install or use a trojan (1c), or gain user-level access to a host. Major
Logging (no recommended
action) Attacks attempt to obtain critical information through directory
traversal or information leaks. Minor
Copyright © 2010, Juniper Networks, Inc. 476
Network and Security Manager Administration Guide