Cisco PIX Firewall and VPN Configuration Guide, Version 6.2 - Controlling Network Access and Use [Cisco PIX Firewall Software]

Table Of Contents

Controlling Network Access and Use

Allowing Server Access with Static NAT

Allowing Inbound Connections

Controlling Outbound Connectivity

Using the Static Command for Port Redirection

Overview

Port Redirection Configuration

Port Redirection Example

Using Authentication and Authorization

Configuring AAA

Configuring RADIUS Authorization

Access Control Configuration Example

Basic Configuration

Authentication and Authorization

Managing Access to Services

Using TurboACL

Overview

Globally Configuring TurboACL

Configuring Individual TurboACLs

Viewing TuboACL Configuration

Downloading Access Lists

Configuring Downloadable ACLs

Downloading a Named Access List

Downloading an Access List Without a Name

Software Restrictions

Simplifying Access Control with Object Grouping

How Object Grouping Works

Using Subcommand Mode

Configuring and Using Object Groups with Access Control

Configuring Protocol Object Groups

Configuring Network Object Groups

Configuring Service Object Groups

Configuring ICMP-Type Object Groups

Nesting Object Groups

Displaying Configured Object Groups

Removing Object Groups

Filtering Outbound Connections

Filtering ActiveX Objects

Filtering Java Applets

Filtering URLs with Internet Filtering Servers

Overview

Identifying the Filtering Server

Configuring Filtering Policy

Filtering Long URLs

Buffering HTTP Replies for Pending URLs

Setting the URL and URL Pending Buffer Size

Viewing Filtering Statistics and Configuration

Configuration Procedure

Controlling Network Access and Use

This chapter describes how to establish and control network connectivity for different applications and implementations after you have completed your basic configuration, described in Chapter 2, "Establishing Connectivity." This chapter contains the following sections:

•Allowing Server Access with Static NAT

•Allowing Inbound Connections

•Controlling Outbound Connectivity

•Using the Static Command for Port Redirection

•Using Authentication and Authorization

•Access Control Configuration Example

•Using TurboACL

•Downloading Access Lists

•Simplifying Access Control with Object Grouping

•Filtering Outbound Connections

Allowing Server Access with Static NAT

Static Network Address Translation (NAT) creates a permanent, one-to-one mapping between an address on an internal network (a higher security level interface) and a perimeter or external network (lower security level interface). For example, to share a web server on a perimeter interface with users on the public Internet, use static address translation to map the server's actual address to a registered IP address. Static address translation hides the actual address of the server from users on the less secure interface, making casual access by unauthorized users less likely. Unlike NAT or PAT, it requires a dedicated address on the outside network for each host, so it does not save registered IP addresses.

If you use a static command to allow inbound connections to a fixed IP address, use the access-list and access-group commands to create an access list and to bind it to the appropriate interface. For more information, refer to "Allowing Inbound Connections."

Note Do not use the PIX Firewall interface address with the static command if Stateful Failover is enabled. Doing this will prevent Stateful Failover from receiving its interface monitoring probes, which run over IP protocol 105, and as a result, the interface will appear to be in waiting state. For further information about Stateful Failover, refer to Chapter 10, "Using PIX Firewall Failover."

The main options of the static command are as follows:
static [(internal_if_name, external_if_name)] global_ip local_ip [netmask network_mask]  
[max_conns]
•Replace internal_if_name with the internal network interface name. In general, this is the higher security level interface you are accessing.

•Replace external_if_name with the external network interface name. In general, this is the lower security level interface you are accessing.

•Replace global_ip with the outside (global) IP address. In general, this is the interface with the lower security level. This address cannot be a PAT IP address.

•Replace local_ip with the internal (local) IP address from the inside network. In general, this is the interface with the higher security level.

•Replace network_mask with the network mask pertains to both global_ip and local_ip. For host addresses, always use 255.255.255.255. For network addresses, use the appropriate subnet mask for the network.

•(Optional) replace max_conns with the maximum number of concurrent connections permitted through the static address translation.

Note To configure static translation for a host residing on the less secure interface (using outside NAT) reverse the interface in the static command. Refer to the Cisco PIX Firewall Command Reference for more information about the static command.

For example, the following command maps a server with an internal IP address of 10.1.1.3 to the registered IP address 209.165.201.12:
static (inside, outside) 209.165.201.12 10.1.1.3 netmask 255.255.255.255 0 0
This command simply maps the addresses; make sure you also configure access using the access-list and access-group commands, as described in the next section. Also, you will must inform the DNS administrator to create an MX record for the external address so that traffic sent to the server host name is directed to the correct address.

Note For information about how to configure static translation without NAT, refer to the static command in the Cisco PIX Firewall Command Reference.

Allowing Inbound Connections

By default, the PIX Firewall denies access to an internal or perimeter (more secure) network from an external (less secure) network. You specifically allow inbound connections by using access lists. Access lists work on a first-match basis, so for inbound access, you must deny first and then permit after.

Note Beginning with PIX Firewall version 5.3, access lists are the preferred method for managing network access. The conduit command was used in earlier versions. Access lists provide improved flexibility and greater ease of use for those familiar with Cisco IOS access control. However, the conduit command is still supported to maintain backward compatibility of configurations written for previous PIX Firewall versions.

You use the access-list and access-group commands to permit access based on source or destination IP address, or by the protocol port number. Use the access-list command to create a single access list entry, and use the access-group command to bind one or more access list entries to a specific interface. Only specify one access-group command for each interface.

Note To allow access only for specific users, set up authentication, as described in "Using Authentication and Authorization."

Before you can set up an access list for a host, set up address translation by using a global or static command. Setting up address translation with the global command is described in Chapter 2, "Establishing Connectivity." Setting up address translation using the static command was described earlier in the previous section "Allowing Server Access with Static NAT."

The format for the access-list command is as follows:
access-list ID action protocol source_address port destination_address port
•Replace ID with a name or number you create to identify a group of access-list command statements; for example, "acl_out," which identifies that the permissions apply to access from the outside interface.

•Replace action with permit or deny depending on whether you want to permit or deny access to the server. By default, all inbound access is denied, so you will must permit access to a specific protocol or port.

•Replace protocol with the protocol (tcp or udp). For most servers, such as HTTP or email, use tcp.

•Replace source_address with the host or network address for those systems on the lower security level interface that must access the destination_address. Use any to let any host access the destination_address. If you specify a single host, precede the address with host; for example host 192.168.1.2. If you specify a network address, also specify a network mask; for example, 192.168.1.0 255.255.255.0.

•Replace the first port parameter with the protocol port used by the source host to initiate the connection.

•Replace destination_address with the host or network global address that you specified with the static command statement. For a host address, precede the address with host; for networks, specify the network address and the appropriate network mask.

•Replace the second port parameter with the literal port name or number for the destination server protocol. For a web server, use the string http or the number 80. For an email server, use smtp or the number 25. The port is preceded with the eq (equals) parameter.

The following is a list of literal port names that you can use when configuring an access-list command statement: DNS, ESP, FTP, H323, HTTP, IDENT, NNTP, NTP, POP2, POP3, PPTP, RPC, SMTP, SNMP, SNMPTRAP, SQLNET, TCP, Telnet, TFTP, and UDP. You can also specify these ports by number. Port numbers are defined in RFC 1700.

Two access-list command statement definitions are required to permit access to the following ports:

–DNS, Discard, Echo, Ident, NTP, RPC, SUNRPC, and Talk each require one definition for TCP and one for UDP.

–PPTP requires one definition for port 1723 on TCP and another for port 0 and GRE protocol.

–TACACS+ requires one definition for port 49 on TCP.

The format for the access-group command is as follows:
access-group ID in interface low_interface
Replace ID with the same identifier that you specified in the access-list command statement.

Replace low_interface with the lower security interface that you specified in the static command statement. This is the interface through which users will access the external (global) address.

The following example illustrates the three commands required to enable access to a web server with the external IP address 209.165.201.12:
static (inside, outside) 209.165.201.12 10.1.1.3 netmask 255.255.255.255 0 0
access-list acl_out permit tcp any host 209.165.201.12 eq www
access-group acl_out in interface outside
This example uses the same static command that was shown in the previous section.

Controlling Outbound Connectivity

By default, all connections initiated on a network with a higher security level are allowed out, and you configure any restrictions required. You can control outbound access by IP address and protocol port, or combine access control with user authentication, as described in "Using Authentication and Authorization." If you are not enforcing restrictions on outbound network traffic, you do not need outbound access lists.

An outbound access list lets you restrict users from starting outbound connections or lets you restrict users from accessing specific destination addresses or networks. Access lists work on a first-match basis, so for outbound access lists, you must permit first and then deny after.

For example, you could restrict some users from accessing websites, permit others access, or restrict one or more users from accessing a specific website. Define access restrictions with the access-list command, and use the access-group command to bind the access-list command statements to an interface.

When creating an outbound access list, the format for the access-list command statement is the same as shown earlier in "Allowing Inbound Connections:"
access-list ID action protocol source_address port destination_address port
By default, outbound access is permitted, so you use the deny action to restrict access when using an outbound access list.

For example, to prevent users on the 192.168.1.0 network on the inside interface from starting connections on the outside interface and permit all others, specify the 192.168.1.0 network address as the source address and the network connected to the outside interface as the destination address. In the example that follows, the network on the outside interface is 209.165.201.0. The access-list and access-group command statements are as follows.
access-list acl_in deny tcp 192.168.1.0 255.255.255.224 209.165.201.0 255.255.255.224
access-list acl_in permit ip any any
access-group acl_in in interface inside
You can use also use access lists to prevent access to a specific server. For example, if you want to restrict users on the inside interface from accessing a website at address 209.165.201.29 on the outside interface (while allowing other outbound access), use the following commands:
access-list acl_in deny tcp any host 209.165.201.29 eq www
access-list acl_in permit ip any any
access-group acl_in in interface inside
These commands let any users start connections, but not to 209.165.201.29. The access-group command specifies that the users are on the inside interface.

Note If controlling outbound access in your network is an important issue, consider using the Websense filtering application, described in "Filtering URLs with Internet Filtering Servers."

Using the Static Command for Port Redirection

This section describes the port redirection feature, introduced in PIX Firewall version 6.0. It includes the following topics:

•Overview

•Port Redirection Configuration

•Port Redirection Example

Overview

Port redirection allows users on a lower security interface to connect to a particular IP address and port and to have the PIX Firewall redirect the traffic to the appropriate server on a higher security interface.

The shared address can be a unique address, a shared outbound PAT address, or an address shared with the external interface. To implement port redirection, use the following command.
static [(internal_if_name, external_if_name)] {tcp|udp} {global_ip|interface} global_port 
local_ip local_port [netmask mask]
For an explanation of this command syntax, refer to the Cisco PIX Firewall Command Reference.

Port Redirection Configuration

Figure 3-1 illustrates a typical network scenario in which the port redirection feature might be useful.

Figure 3-1 Port Redirection Using the Static Command

In the configuration described in this section, port redirection occurs for users on external networks as follows:

•Telnet requests to unique IP address 209.165.201.5 are redirected to 10.1.1.6

•FTP requests to unique IP address 209.165.201.5 are redirected to 10.1.1.3

•Telnet requests to PAT address 209.165.201.15 are redirected to 10.1.1.4

•Telnet requests to the PIX Firewall outside IP address 209.165.201.25 are redirected to 10.1.1.5

•HTTP request to PIX Firewall outside IP address 209.165.201.25 are redirected to 10.1.1.5

•HTTP port 8080 requests to PAT address 209.165.201.15 are redirected to 10.1.1.7 port 80

To implement this scenario, complete the following steps:

Step 1 Configure application inspection of FTP requests on port 21 by entering the following command:
fixup protocol ftp 21
Step 2 Configure the IP address of the lower and higher security interfaces of your PIX Firewall by entering the following command:
ip address outside 209.165.201.25 255.255.255.0
ip address inside 10.1.1.2 255.255.255.0
Step 3 Identify a global PAT address for the lower security interface by entering the following command:
global (outside) 1 209.165.201.15
Step 4 Configure NAT and PAT by entering the following command:
nat (inside) 1 0.0.0.0 0.0.0.0 0 0
Step 5 Redirect Telnet requests for 209.165.201.5:
static (inside,outside) tcp 209.165.201.5 telnet 10.1.1.6 telnet netmask 255.255.255.255 0 
0
This command causes Telnet requests to be redirected to 10.1.1.6.

Step 6 Redirect FTP requests for IP address 209.165.201.5:
static (inside,outside) tcp 209.165.201.5 ftp 10.1.1.3 ftp netmask 255.255.255.255 0 0
This command causes FTP requests to be redirected to 10.1.1.3.

Step 7 Redirect Telnet requests for PAT address 209.165.201.15:
static (inside,outside) tcp 209.165.201.15 telnet 10.1.1.4 telnet netmask 255.255.255.255 
0 0
This command causes Telnet requests to be redirected to 10.1.1.4.

Step 8 Redirect Telnet requests for the PIX Firewall outside interface address:
static (inside,outside) tcp interface telnet 10.1.1.5 telnet netmask 255.255.255.255 0 0
This command causes Telnet requests to be redirected to 10.1.1.5.

Step 9 Redirect HTTP requests for the PIX Firewall outside interface address:
static (inside,outside) tcp interface www 10.1.1.5 www netmask 255.255.255.255 0 0
This command causes HTTP request to be redirected to 10.1.1.5.

Step 10 Redirect HTTP requests on port 8080 for PAT address 209.165.201.15:
static (inside,outside) tcp 209.165.201.15 8080 10.1.1.7 www netmask 255.255.255.255 0 0
This command causes HTTP port 8080 requests to be redirected to 10.1.1.7 port 80.

Port Redirection Example

Example 3-1 illustrates the configuration required to implement the port redirection described in this scenario.

Example 3-1 Port Redirection with the static Command
fixup protocol ftp 21
ip address outside 209.165.201.25 255.255.255.0
ip address inside 10.1.1.2 255.255.255.0
global (outside) 1 209.165.201.15
nat (inside) 1 0.0.0.0 0.0.0.0 0 0
static (inside,outside) tcp 209.165.201.5 telnet 10.1.1.6 telnet netmask 255.255.255.255 0 
0
static (inside,outside) tcp 209.165.201.5 ftp 10.1.1.3 ftp netmask 255.255.255.255 0 0
static (inside,outside) tcp 209.165.201.15 telnet 10.1.1.4 telnet netmask 255.255.255.255 
0 0
static (inside,outside) tcp interface telnet 10.1.1.5 telnet netmask 255.255.255.255 0 0
static (inside,outside) tcp interface www 10.1.1.5 www netmask 255.255.255.255 0 0
static (inside,outside) tcp 209.165.201.15 8080 10.1.1.7 www netmask 255.255.255.255 0 0
Using Authentication and Authorization

You can use access lists to control traffic based on IP address and protocol, but to control access and use for specific users or groups, you must use authentication and authorization. Authentication, which is the process of identifying users, is supported by the PIX Firewall for RADIUS and TACACS+ servers. Authorization identifies the specific permissions for a given user.

If you want to apply authentication and authorization when an internal (local) host initiates a connection to an external (lower security) network, enable it on the internal (higher security) interface. To set up authentication and authorization to occur when an external host initiates a connection to an internal host, enable it on the outside interface.

Note If you want a host on an outside (lower security level) interface to initiate connections with a host on an internal (higher security level) interface, create static and access-list command statements for the connection.

This section includes the following topics:

•Configuring AAA

•Configuring RADIUS Authorization

Configuring AAA

To enable authentication and authorization, identify the authentication server you are using and the server encryption key on the PIX Firewall. From the configuration on the authentication server you must determine the users that can access the network, the services that they can use, and the hosts that they can access. Once you have this information, you can configure the PIX Firewall to either enable or disable authentication or authorization.

In addition, you can configure the PIX Firewall to control user access to specific hosts or services. However, it is easier to maintain this kind of access control in a single location, at the authentication server. After you enable authentication and authorization, the PIX Firewall provides prompts inbound or outbound for users of FTP, Telnet, or HTTP (Web) access. Controlling access to a specific system or service is handled by the authentication and authorization server.

Follow these steps to enable the PIX Firewall to support TACACS+ user authentication and authorization:

Step 1 For inbound authentication, create the static and access-list command statements required to permit outside hosts to access servers on the inside network.

Step 2 If the internal network connects to the Internet, create a global address pool of registered IP addresses. Then specify the inside hosts that can start outbound connections with the nat command and with the access control lists features found in the outbound and apply commands.

Step 3 Identify the server that handles authentication or authorization using the aaa-server command. Create a unique server group name.

For example:
aaa-server AuthInbound protocol tacacs+
aaa-server AuthInbound (inside) host 10.1.1.1 TheUauthKey 
aaa-server AuthOutbound protocol tacacs+
aaa-server AuthOutbound (inside) host 10.1.1.2 TheUauthKey 
The first command statement creates the AuthInbound authentication group using TACACS+ authentication. The second command statement states that the AuthInbound server is on the inside interface, that its IP address is 10.1.1.1, and the encryption key is "TheUauthKey."

The third command statement creates the AuthOutbound authentication group using TACACS+ authentication. The fourth command statement states that the AuthOutbound server is on the inside interface, that its IP address is 10.1.1.2, and the encryption key is "TheUauthKey."

Note RADIUS authorization is provided with the access-list command statement as described in "Configuring RADIUS Authorization."

Step 4 Enable authentication with the aaa authentication command:
aaa authentication include ftp outbound 0 0 0 0 AuthOutbound
aaa authentication include telnet outbound 0 0 0 0 AuthOutbound
aaa authentication include http outbound 0 0 0 0 AuthOutbound
aaa authentication include ftp inbound 0 0 0 0 AuthInbound
aaa authentication include telnet inbound 0 0 0 0 AuthInbound
aaa authentication include http inbound 0 0 0 0 AuthInbound
The AuthInbound and AuthOutbound groups are those you specified with the aaa-server command.

Note Be careful to apply authentication only to protocols that can be authenticated. Applying authentication using the any keyword will prevent protocols such as SMTP or HTTPS from passing through the PIX Firewall.

Step 5 Enable authorization with the aaa authorization command. PIX Firewall checks the authorization request with the AAA server, which makes the decision about what services a user can access. Use one or both of the following commands to specify outbound and inbound authorization.
aaa authorization include ftp outbound 0 0 0 0
aaa authorization include telnet outbound 0 0 0 0
aaa authorization include http outbound 0 0 0 0
aaa authorization include ftp inbound 0 0 0 0
aaa authorization include telnet inbound 0 0 0 0
aaa authorization include http inbound 0 0 0 0
You can specify port ranges for the aaa authorization command in the following format:
aaa authorization include | exclude author_service inbound | outbound | if_name local_ip 
local_mask foreign_ip foreign_mask
where:

•author_service—The service that PIX Firewall listens to for AAA connections. Possible values are any, http, ftp, telnet, or protocol/port. Use the IP protocol number for protocol or use the TCP/UDP destination port or port range. A port value of 0 means all ports. You can also use a specific ICMP type.

•inbound, outbound, if_name—Specify whether users are authenticated and authorized on inbound or outbound connections, or for connections that arrive at a specific interface.

•local_ip, local_mask—Specify the IP address on the higher security level interface from which or to which access is required.

•foreign_ip, foreign_mask—Specify the IP address on the lower security level interface from which or to which access is required.

Configuring RADIUS Authorization

PIX Firewall allows a RADIUS server to send user group attributes to the PIX Firewall in the RADIUS authentication response message.

The administrator first defines access lists on the PIX Firewall for each user group. For example, there could be access lists for each department in an organization, sales, marketing, engineering, and so on. The administrator then lists the access list in the group profile in CiscoSecure.

After the PIX Firewall authenticates a user, it can then use the RADIUS attribute (attribute 11, filter-id) returned by the authentication server to identify an access list for a given user group. To maintain consistency, PIX Firewall also provides the same functionality for TACACS+.

Note Access lists can be used with either RADIUS or TACACS but authorizing FTP, HTTP, or Telnet is only possible with TACACS+.

To restrict users in a department to three servers and deny everything else, the access-list command statements are as follows:
access-list eng permit ip any server1 255.255.255.255
access-list eng permit ip any server2 255.255.255.255
access-list eng permit ip any server3 255.255.255.255
access-list eng deny ip any any
In this example, the vendor-specific attribute string in the CiscoSecure configuration has been set to acl=eng. Use this field in the CiscoSecure configuration to identify the access-list identification name. The PIX Firewall gets the acl=acl_ID from CiscoSecure and extracts the ACL number from the attribute string, which it puts in a user's uauth entry. When a user tries to open a connection, PIX Firewall checks the access list in the user's uauth entry, and depending on the permit or deny status of the access list match, permits or denies the connection. When a connection is denied, PIX Firewall generates a corresponding syslog message. If there is no match, then the implicit rule is to deny.

Because the source IP of a given user can vary depending on where they are logging in from, set the source address in the access-list command statement to any, and the destination address to identify the network services to which user is permitted or denied access.

The aaa authorization command does not require or provide a separate RADIUS option. To enable RADIUS authorization, perform the following steps:

Step 1 Enable RADIUS authentication with the aaa authentication command.

Step 2 Create the access-list command statements to specify the services that hosts are authorized to use with RADIUS in the PIX Firewall.

Step 3 Configure the authentication server with the vendor-specific acl=acl_ID identifier to specify the access-list ID.

When the PIX Firewall sends a request to the authentication server, it returns the acl=acl_ID string, which tells PIX Firewall to use the access-list command statements to determine how RADIUS users are authorized.

Access Control Configuration Example

This section provides an example of how to implement access control and includes the following topics:

•Basic Configuration

•Authentication and Authorization

•Managing Access to Services

Basic Configuration

Figure 3-2 illustrates the network configuration used in this example.

Figure 3-2 Two Interfaces with NAT—Access Control

The following procedure shows the basic configuration required for this example. This procedure is similar to the configuration shown in "Configuration Examples," in Chapter 2, "Establishing Connectivity":

Step 1 Identify the security level and names of each interface by entering the following commands:
nameif ethernet0 outside security0
nameif ethernet1 inside security100
Step 2 Identify the line speed of each interface by entering the following commands:
interface ethernet0 10baset
interface ethernet1 10baset
You may get better performance by changing the default auto option in the interface command to the specific line speed for the interface card.

Step 3 Identify the IP addresses for each interface:
ip address inside 10.1.1.1 255.255.255.0
ip address outside 209.165.201.1 255.255.255.224
Step 4 Specify the host name for the PIX Firewall:
hostname pixfirewall
This name appears in the command line prompt.

Step 5 Let inside IP addresses be recognized on the outside network and let inside users start outbound connections:
nat (inside) 1 0.0.0.0 0.0.0.0
nat (inside) 2 192.168.3.0 255.255.255.0
global (outside) 1 209.165.201.6-209.165.201.8 netmask 255.255.255.224
global (outside) 1 209.165.201.10 netmask 255.255.255.224
global (outside) 2 209.165.200.225-209.165.200.254 netmask 255.255.255.224
Step 6 Set the outside default route to the router attached to the Internet:
route outside 0 0 209.165.201.4 1
Example 3-2 shows the basic configuration required to implement a PIX Firewall with two interfaces with NAT.

Example 3-2 Two Interfaces with NAT—Basic Configuration
nameif ethernet0 outside security0
nameif ethernet1 inside security100
interface ethernet0 10baset
interface ethernet1 10baset
ip address inside 10.1.1.1 255.255.255.0
ip address outside 209.165.201.1 255.255.255.224
hostname pixfirewall
nat (inside) 1 0.0.0.0 0.0.0.0
nat (inside) 2 192.168.3.0 255.255.255.0
global (outside) 1 209.165.201.6-209.165.201.8 netmask 255.255.255.224
global (outside) 1 209.165.201.10 netmask 255.255.255.224
global (outside) 2 209.165.200.225-209.165.200.254 netmask 255.255.255.224
route outside 0 0 209.165.201.4 1
Authentication and Authorization

This section describes how to implement authentication and authorization for traffic through the PIX Firewall, using a TACACS+ server. The commands used for this purpose are in addition to the basic firewall configuration required, which is described in the previous section, "Basic Configuration."

The aaa-server command specifies the IP address of the TACACS+ authentication server. The aaa authentication command statement specifies that users on network 192.168.3.0 starting FTP, HTTP, and Web connections from the inside interface be prompted for their usernames and passwords before being permitted to access the servers on other interfaces. The aaa authorization command statement lets the users on 192.168.3.0 access FTP, HTTP, or Telnet, and any TCP connections to anywhere as authorized by the AAA server. Even though it appears that the aaa commands let the PIX Firewall set security policy, the authentication server actually does the work to decide which users are authenticated and what services they can access when authentication is permitted.

Example 3-3 shows the command listing for configuring access to services for the network illustrated in Figure 3-2.

Example 3-3 Authentication and Authorization Commands
aaa-server TACACS+ (inside) host 10.1.1.12 1q2w3e
aaa authentication include ftp inside 192.168.3.0  255.255.255.0 0 0 TACACS+
aaa authorization include ftp inside 192.168.3.0 255.255.255.0 0 0
aaa authentication include http inside 192.168.3.0       255.255.255.0 0 0 TACACS+
aaa authorization include http inside 192.168.3.0 255.255.255.0 0 0
aaa authentication include telnet inside 192.168.3.0       255.255.255.0 0 0 TACACS+
aaa authorization include telnet inside 192.168.3.0 255.255.255.0 0 0
Managing Access to Services

Note The commands in this section are used in addition to the basic firewall configuration required, which is described in the previous section, "Basic Configuration."

The following procedure shows the commands required to manage user access to H.323 and Web services:

Step 1 Create outbound access lists to determine which hosts can access services:
access-list acl_in deny tcp host 192.168.3.3 any eq 1720 
access-list acl_in deny tcp any any eq 80 
access-list acl_in permit host 192.168.3.3 any eq 80 
access-list acl_in permit host 10.1.1.11 any eq 80 
The first access-list command statement denies host 192.168.3.3 from accessing H.323 (port 1720) services such as MS NetMeeting or Intel Internet Phone. The next command statement denies all hosts from accessing the Web (port 80). The next command statement permits host 192.168.3.3 to use the Web. The last access-list command statement permits host 10.1.1.11 access to the Web (at port 80).

Step 2 Specify that the access-list group regulates the activities of inside hosts starting outbound connections:
access-group acl_in interface inside
Step 3 Create static address mappings:
static (inside, outside) 209.165.201.16 192.168.3.16 netmask 255.255.255.240 
The static command statement creates a net static command statement, which is a static command statement for a set of IP addresses, in this case for IP addresses 209.165.201.17 through 209.165.201.30.

Step 4 Enable VoIP access:
access-list acl_out permit tcp any host 209.165.201.16 eq h323
The access-list command statement lets users on the Internet send Intel Internet Phone (port h323) requests to users on 192.168.3.x while addressing them as 209.165.201.x.

Step 5 Establish an externally visible IP address for Web access:
static (inside, outside) 209.165.201.11 10.1.1.11 
access-list acl_out permit tcp any host 209.165.201.11 eq 80 
The static command statement with the access-list command statement establishes an externally visible IP address for Web access (port 80 in the access-list command statement).

Example 3-4 shows the command listing for configuring access to services for the network illustrated in Figure 3-2.

Example 3-4 Configuring Access to Services
access-list acl_in deny tcp host 192.168.3.3 any eq 1720 
access-list acl_in deny tcp any any eq 80 
access-list acl_in permit host 192.168.3.3 any eq 80 
access-list acl_in permit host 10.1.1.11 any eq 80 
access-group acl_in interface inside
static (inside, outside) 209.165.201.16 192.168.3.16 netmask 255.255.255.240 
access-list acl_out permit tcp any host 209.165.201.16 eq h323
static (inside, outside) 209.165.201.11 10.1.1.11 
access-list acl_out permit tcp any host 209.165.201.11 eq 80 
Using TurboACL

This section describes how to use the TurboACL feature introduced with PIX Firewall version 6.2. It includes the following topics:

•Overview

•Globally Configuring TurboACL

•Configuring Individual TurboACLs

•Viewing TuboACL Configuration

Overview

An access list typically consists of multiple access list entries, organized internally by PIX Firewall as a linked list. When a packet is subjected to access list control, the PIX Firewall searches this linked list linearly to find a matching element. The matching element is then examined to determine if the packet is to be transmitted or dropped. With a linear search, the average search time increases proportional to the size of the list.

TurboACL is a feature introduced with PIX Firewall version 6.2 that improves the average search time for access control lists containing a large number of entries. The TurboACL feature causes the PIX Firewall to compile tables for ACLs and this improves searching of long ACLs.

You can enable this feature for the entire PIX Firewall and then disable it for specific ACLs, or enable it only for specific ACLs. For short ACLs, TurboACL does not improve performance. A TurboACL search, no matter how short the ACL, requires about the same amount of time as a regular ACL search of from twelve to eighteen entries. For this reason, even when enabled, the TurboACL feature is only applied to ACLs with nineteen or more entries.

The TurboACL feature requires significant amounts of memory and is most appropriate for high-end PIX Firewall models, such as the PIX 525 or PIX 535. The minimum memory required for TurboACL is 2.1 MB and approximately 1 MB of memory is required for every 2000 ACL elements.

Note With PIX Firewall models having limited memory, such as the PIX 501, implementing the TurboACL feature may cause problems, such as not being able to load Cisco PIX Device Manager (PDM). If memory problems occur after enabling TurboACL, disable it using the no access-list compiled command.

The actual amount of memory required depends not only on the number of ACL elements but also on the complexity of the entries.

Note When you add or delete an element from a turbo-enabled ACL, the internal data tables associated with the ACL are regenerated, and this produces an appreciable load on the PIX Firewall CPU.

Globally Configuring TurboACL

The syntax for enabling TurboACL for the entire PIX Firewall is as follows:

[no] access-list compiled
This configures TurboACL on all ACLs having 19 or more entries. This command causes the TurboACL process to scan through all existing ACLs. During the scan, it marks and turbo-compiles any ACL which has 19 or more Access Control Entries (ACEs) and has not yet been turbo-compiled.

The command no access-list compiled, which is the default, causes the TurboACL process to scan through all compiled ACLs and mark every one as non-turbo. It also deletes all existing TurboACL structures.

When the PIX Firewall is running, the command access-list compiled marks every ACL to be turbo-configured, and the command no access-list compiled marks every ACL as non-turbo.

Configuring Individual TurboACLs

The individual TurboACL command can be used to enable individual turbo configuration for individual ACLs when TurboACL is not globally enabled. Also, after globally configuring TurboACL, you can disable individual ACL's turbo-configuration by using the individual TurboACL command. The syntax of this command is as follows.
[no] access-list acl_name compiled
This command is used to individually enable or disable TurboACL on a specific ACL. The acl_name must specify an existing ACL.This command will cause the TurboACL process to mark the ACL specified by acl_name as turbo-configured, and to turbo-compile the ACL if the ACL has 19 or more ACEs and has not yet been turbo-compiled.

If you enter the no form of the command, the TurboACL process deletes the TurboACL structures associated with the ACL and marks the ACL as non-turbo.

Viewing TuboACL Configuration

The show access-list command displays the memory usage of each individually turbo-compiled ACL and the shared memory usage for all the turbo-compiled ACLs. If no ACL is turbo-compiled, no turbo-statistic is displayed. This command also shows the number of ACEs in an ACL and whether an ACL is configured with TurboACL. Note that an ACL may be configured for turbo but it will not be compiled unless the number of ACEs exceeds the threshold. In such a case, this command will show that the ACL is turbo-configured, but there will not be any entry for the ACL in the TurboACL statistic output.

Example 3-5 provides an example of the output of the show access-list command:

Example 3-5 TurboACL Statistics
pix# show access list
TurboACL statistics:
ACL                   State       Memory(KB)
--------------------- ----------- ----------
acl_foo               Operational   5         
Acl_bar               Operational   2         
Shared memory usage: 2046 KB
access-list compiled
access-list acl_foo turbo-configured; 19 elements
access-list acl_foo permit tcp any host 11.0.0.252 (hitcnt=0) 
access-list acl_foo permit tcp any host 11.0.0.255 (hitcnt=0) 
access-list acl_foo permit tcp any host 11.0.0.253 (hitcnt=0) 
access-list acl_foo permit tcp 13.0.0.0 255.0.0.0 host 11.0.0.254 eq telnet (hitcnt=2) 
access-list acl_foo permit tcp 13.0.0.0 255.0.0.0 host 11.0.0.254 eq 1 (hitcnt=0) 
Downloading Access Lists

PIX Firewall supports per-user access list authorization, by which a user is authorized to do only what is permitted in the user's individual access list entries. This section describes how to implement this feature and includes the following topics:

•Configuring Downloadable ACLs

•Downloading a Named Access List

•Downloading an Access List Without a Name

•Software Restrictions

Configuring Downloadable ACLs

This feature lets you configure per-user access lists on a AAA server and then download the access list to a PIX Firewall during user authentication.

With PIX Firewall version 6.2, these access lists can be downloaded from a AAA server and do not need to be configured separately on the PIX Firewall. This feature improves scalability when using access lists for individual users.

Note Downloadable ACLs are only supported with RADIUS servers and not with TACACS+ servers.

The following are the two methods for downloading an access list from an AAA server to the PIX Firewall:

•Downloading a named access list—Configure a user (real) authentication profile to include a Shared Profile Component (SPC) and then configure the SPC to include the access list name and the actual access list. This method should be used when there are frequent requests for downloading a large access list.

•Downloading an access list without a name—Configure a user authentication profile on an AAA server to include the PIX Firewall access list to be downloaded. This method should be used when there are no frequent requests for the same access list.

Downloading a Named Access List

To download a named access list during a user authentication, the following configuration must be used on Cisco Secure ACS 3.0 or later:

1. Select Downloadable PIX ACLs from the Shared Profile Component (SPC) menu item.

2. Click Add to add an ACL definition and enter the name, description, and the actual definitions for the ACL.

The ACL definition consists of one or more PIX Firewall access-list commands with each command on a separate line. Each command must be entered without the access-list keyword and the name for the access list because they are not needed. The rest of the command line must conform to the syntax and semantics rules of the PIX Firewall access-list command. A PIX Firewall Syslog message will be logged if there is an error in a downloaded access-list command.

The following is an example of an ACL definition after it is downloaded to the PIX Firewall:
  +--------------------------------------------+
  | Shared profile Components                  |
  |                                            |
  |   Downloadable PIX ACLs                    |
  |                                            |
  | Name:			acs_ten_acl            |
  | Description:  10 PIX access-list commands  |
  |                                            |
  |                                            |
  |      ACL Definitions                       |
  |                                            |
  | permit tcp any host 11.0.0.254             |
  | permit udp any host 11.0.0.254             |
  | permit icmp any host 11.0.0.254            |
  | permit tcp any host 11.0.0.253             |
  | permit udp any host 11.0.0.253             |
  | permit icmp any host 11.0.0.253            |
  | permit tcp any host 11.0.0.252             |
  | permit udp any host 11.0.0.252             |
  | permit icmp any host 11.0.0.252            |
  | permit ip any any                          |
  +--------------------------------------------+
On the PIX Firewall:
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit tcp any host 11.0.0.254
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit udp any host 11.0.0.254
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit icmp any host 11.0.0.254
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit tcp any host 11.0.0.253
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit udp any host 11.0.0.253
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit icmp any host 11.0.0.253
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit tcp any host 11.0.0.252
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit udp any host 11.0.0.252
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit icmp any host 11.0.0.252
  access-list #ACSACL#-PIX-acs_ten_acl-3b5385f7 permit ip any any
3. Configure a Cisco Secure ACS user or a group through User Setup or Group Setup to include the defined ACL in the user or group settings.

Once the configuration is properly configured, a user authentication request will first cause the access list name to be sent to the PIX Firewall. The PIX Firewall will determine if the named ACL already exists and if not, the PIX Firewall will request the ACL to be downloaded. A named, downloadable is not downloaded again as long as it exists on the PIX Firewall.

If the download is successful, the ACL on the PIX Firewall will have the following name:

#ACSACL#-acl_name-12345678

Where acl_name is the name for the access list defined in the SPC and 12345678 is a unique version ID. If the named access list is not configured on ACS or the download fails for any other reason, a Syslog message will be logged.

4. Activate the use of downloadable ACLs by performing the following steps:

a. Select Interface Configuration from the Cisco Secure ACS main menu.

b. Select Advanced Options from the Interface Configuration menu.

c. Check either or both of the following options:

–User-Level Downloadable ACLs

–Group-Level Downloadable ACLs

Downloading an Access List Without a Name

To download an access list without using a name during a user authentication, perform the following steps at a AAA RADIUS server:

1. Configure CISCO-specific VSA (Attribute 26) string of a user authentication profile in the following format:
ip:inacl#nnn=ACL_COMMAND
where:

•ip:inacl# is the string that specifies an input ACL.

•nnn is a number in the range from 0 to 999999999 that identifies the order of the access-list command statement to be configured on the PIX Firewall. If this parameter is omitted the sequence value is 0.

•ACL_COMMAND represents one ore more PIX Firewall access-list commands.

Statements are separated by colons (:). Statements should not include the access-list command or the access list name. You can configure multiple occurrences of the string "ip:inacl#nnn=" in the same user authentication profile to define a PIX Firewall access list. If multiple entries have the same sequence number, they will be configured in the same order as they appear in the Cisco-specific VSA attribute.

Multiple lines may be used to configure multiple elements, but an element must be completely contained on a single line. For example, the permit tcp any any command cannot be broken into two separate lines.

A downloadable ACL without a name is assigned a name by the PIX Firewall after it is downloaded in the following format:
AAA-user-username
Where username is the name of the user that is being authenticated.

If an access-list command statement has a syntax or semantics error, or if the no access-list command is used (an empty access-list), Syslog messages will be generated. However, an error with a single access-list command does not abort the processing of the entire downloaded ACL.

Example 3-6 Example Configuration for a Downloadable Access List

The following configuration would be entered for the user Admin in the [009\001] cisco-av-pair field under Group Setup>Cisco IOS/PIX RADIUS Attributes:
ip:inacl#1=permit tcp 13.0.0.0 255.0.0.0 11.0.0.0 255.0.0.0
ip:inacl#99=deny tcp any any
ip:inacl#2=permit udp 13.0.0.0 255.0.0.0 11.0.0.0 255.0.0.0
ip:inacl#99=deny udp any any
ip:inacl#3=permit icmp 13.0.0.0 255.0.0.0 11.0.0.0 255.0.0.0
The resulting downloaded access-list commands on PIX Firewall are as follows:
access-list AAA-user-foo; 5 elements
access-list AAA-user-foo permit tcp 13.0.0.0 255.0.0.0 11.0.0.0 255.0.0.0
access-list AAA-user-foo permit udp 13.0.0.0 255.0.0.0 11.0.0.0 255.0.0.0
access-list AAA-user-foo permit icmp 13.0.0.0 255.0.0.0 11.0.0.0 255.0.0.0
access-list AAA-user-foo deny tcp any any
access-list AAA-user-foo deny udp any any
Software Restrictions

When downloading access lists via RADIUS, the following restrictions apply:

•RADIUS packet size is limited to 4096, but the actual size for the access list can vary considerably depending on the existence of other variable-length, RADIUS attributes.

•Each RADIUS attribute field used to specify access lists is limited to 253 bytes.

Note If there exists any incompatibility between the PIX Firewall and the Cisco IOS access list, the incompatibility will also exist for the downloaded access list. In other words, an access list defined for PIX Firewall on a AAA server may not be valid if the access list is downloaded to Cisco IOS software, and vice versa.

Simplifying Access Control with Object Grouping

This section describes how to use Object Grouping, a feature introduced in PIX Firewall version 6.2, for simplifying complex access control policies. It includes the following topics:

•How Object Grouping Works

•Using Subcommand Mode

•Configuring and Using Object Groups with Access Control

•Configuring Protocol Object Groups

•Configuring Network Object Groups

•Configuring Service Object Groups

•Configuring ICMP-Type Object Groups

•Nesting Object Groups

•Displaying Configured Object Groups

•Removing Object Groups

How Object Grouping Works

Object Grouping provides a way to reduce the number of access rules required to describe complex security policies. An access rule can apply to the following types of objects:

•Client host—Makes HTTP, Telnet, FTP, Voice over IP, and other service requests

•Server host—Responds to service requests

•Service type—Services are assigned to well-known, dynamically assigned, or secondary channel TCP or UDP ports

•Subnet—The network address of internal or external subnetworks where server or client hosts are located

•ICMP types—Such as ECHO-REPLY

An access rule allows or denies traffic matching a specific combination of these objects. For example, an access rule might cause the PIX Firewall to allow a designated client to access a particular server host for a specific service. When there is only one client, one host, and one service, only one access rule is needed. However, as the number of clients, servers, and services increases, the number of rules required may increase exponentially.

Object Grouping provides a way to group objects of a similar type into a group so that a single access rule can apply to all the objects in the group. For example, consider the following three object groups:

•MyServices—Includes the TCP/UDP port numbers of the service requests that are allowed access to the internal network

•TrustedHosts—Includes the host and network addresses allowed access to the greatest range of services and servers

•PublicServers—Includes the host addresses of servers to which the greatest access is provided

After creating these groups, you could use a single access rule to allow trusted hosts to make specific service requests to a group of public servers. Object groups can also contain other object groups or be contained by other object groups.

Object Grouping dramatically compresses the number of access rules required to implement a particular security policy. For example, a customer policy that required 3,300 access rules might only require 40 rules after hosts and services are properly grouped.

Using Subcommand Mode

The general syntax of the object-group command is as follows:
[no] object-group object-type grp-id
Replace object-type with one of the following object types:

•protocol—Group of IP protocols. It can be one of the keywords icmp, ip, tcp, or udp, or an integer in the range 1 to 254 representing an IP protocol number. To match any Internet protocol, including ICMP, TCP, and UDP, use the keyword ip.

•service—Group of TCP or UDP port numbers assigned to a different services.

•icmp-type—Group of ICMP message types to which you permit or deny access.

•network—Group of hosts or subnets

Replace grp-id with a descriptive name for the group.

When you enter the object-group command, the prompt changes to the subcommand mode appropriate for the type of object. Commands entered in the subcommand mode apply to the object type and group name identified in the object-group command.

The prompts in each subcommand mode are as follows:
pix(config-protocol)#
pix(config-service)#
pix(config-icmp-type)#
pix(config-network)#