Table Of Contents
Supported Multimedia Applications
Supported Protocols and Applications
TCP/IP Reference Information
This appendix includes the following sections:
IP Addresses
IP address classes are defined as follows:
•
Class A— If the first octet is between 1 and 127 (inclusive), the address is a Class A address. In a Class A address, the first octet is the one-byte net address and the last three octets are the host address. The network mask for Class A addresses is 255.0.0.0.
•
Class B— If the first octet is between 128 and 191 (inclusive), the address is a
Class B address. In a Class B address, the first two octets are the net address and the last two octets are the host address. The network mask for Class B addresses is 255.255.0.0.•
Class C— If the first octet is 192 or higher, the address is a Class C address. In a
Class C address, the first three octets are the net address and the last octet is the host address. The network mask for Class C addresses is 255.255.255.0.We recommend that you use RFC 1918 IP addresses for inside and perimeter addresses. These addresses follow:
•
Class A: 10.0.0.0 to 10.255.255.255
•
Class B: 172.16.0.0 to 172.31.255.255
•
Class C: 192.168.0.0 to 192.168.255.255
PIX Firewall requires that IP addresses in the ip address, static, global, failover, and virtual commands be unique. These IP addresses cannot be the same as your router IP addresses.
In this guide, the use of "address" and "IP address" are synonymous.
IP addresses are primarily one of these values:
•
local_ip—An untranslated IP address on the internal, protected network. In an outbound connection originated from local_ip, the local_ip is translated to the global_ip. On the return path, the global_ip is translated to the local_ip. The local_ip to global_ip translation can be disabled with the nat 0 0 0 command. In syslog messages, this address is referenced as laddr.
•
global_ip—A translated global IP address in the pool or those addresses declared with the global or static commands. In syslog messages, this address is referenced as gaddr.
•
foreign_ip—An untranslated IP address on an external network. foreign_ip is an address for hosts on the external network. If the alias command is in use, an inbound message originating for the foreign_ip source address is translated to dnat_ip by PIX Firewall.
•
dnat_ip—(dual NAT) A translated (by the alias command) IP address on an external network. In an outbound connection destined to dnat_ip, it will be untranslated to foreign_ip. In syslog messages, this address is referenced as faddr.
•
virtual_ip—(used with the virtual command) A fictitious public or private IP address that is not the address of a real web server on the interface you are accessing. We recommend that you use an RFC 1918 address or one you make up.
Ports
The following literal names can be used instead of a numerical port value in command lines:
PIX Firewall permits the following TCP literal names: bgp, chargen, cmd, daytime, discard, domain, echo, exec, finger, ftp, ftp-data, gopher, h323, hostname, http, ident, irc, klogin, kshell, lpd, nntp, pop2, pop3, pptp, rpc, smtp, sqlnet, sunrpc, tacacs, talk, telnet, time, uucp, whois, and www.
Note
PIX Firewall uses port 1521 for SQL*Net. This is the default port used by Oracle for SQL*Net; however, this value does not agree with IANA port assignments.
Note
PIX Firewall listens for RADIUS on ports 1645 and 1646. If your RADIUS server uses ports 1812 and 1813, you will need to reconfigure it to listen on ports 1645 and 1646.
Permitted UDP literal names are biff, bootpc, bootps, discard, dnsix, echo, mobile-ip, nameserver, netbios-dgm, netbios-ns, ntp, rip, snmp, snmptrap, sunrpc, syslog, tacacs, talk, tftp, time, who, and xdmcp.
Note
To assign a port for DNS access, use domain, not dns. The dns keyword translates into the port value for dnsix.
Port numbers can be viewed online at the IANA website:
http://www.iana.org/assignments/port-numbers
Table D-1 lists the literal values.
Protocols and Applications
Possible literal values are ahp, eigrp, esp, gre, icmp, igmp, igrp, ip, ipinip, ipsec, nos, ospf, pcp, snp, tcp, and udp. You can also specify any protocol by number. The esp and ah protocols only work in conjunction with Private Link.
Note
PIX Firewall does not pass multicast packets. Many routing protocols use multicast packets to transmit their data. If you need to send routing protocols across the PIX Firewall, configure the routers with the Cisco IOS software neighbor command. We consider it inherently dangerous to send routing protocols across the PIX Firewall. If the routes on the unprotected interface are corrupted, the routes transmitted to the protected side of the firewall will pollute routers there as well.
Table D-2 lists the numeric values for the protocol literals.
Protocol numbers can be viewed online at the IANA website:
http://www.iana.org/assignments/protocol-numbers
Supported Multimedia Applications
PIX Firewall supports the following multimedia and video conferencing applications:
•
CUseeMe Networks CU-SeeMe
•
CUseeMe Networks CU-SeeMe Pro
•
CUseeMe Networks MeetingPoint
•
Intel Internet Video Phone
•
Microsoft NetMeeting
•
Microsoft NetShow
•
NetMeeting
•
RealNetworks RealAudio and RealVideo
•
VDOnet VDOLive
•
VocalTec Internet Phone
•
VXtreme WebTheater
•
Xing StreamWorks
Supported Protocols and Applications
PIX Firewall supports the following TCP/IP protocols and applications:
•
Address Resolution Protocol (ARP)
•
Archie
•
Berkeley Standard Distribution (BSD)-rcmds
•
Bootstrap Protocol (BOOTP)
•
Domain Name System (DNS)
•
File Transfer Protocol (FTP)
•
Generic Route Encapsulation (GRE)
•
Gopher
•
HyperText Transport Protocol (HTTP)
•
Internet Control Message Protocol (ICMP)
•
Internet Protocol (IP)
•
NetBIOS over IP (Microsoft Networking)
•
Point-to-Point Tunneling Protocol (PPTP)
•
Simple Network Management Protocol (SNMP)
•
Sitara Networks Protocol (SNP)
•
SQL*Net (Oracle client/server protocol)
•
Sun Remote Procedure Call (RPC) services, including Network File System (NFS)
•
Telnet
•
Transmission Control Protocol (TCP)
•
Trivial File Transfer Protocol (TFTP)
•
User Datagram Protocol (UDP)
Using Subnet Masks
This section lists information by subnet mask and identifies which masks are for networks, hosts, and broadcast addresses.
Note
In some networks, broadcasts are also sent on the network address.
This section includes the following topics:
Masks
For the PIX Firewall commands that accept network masks, specify the correct mask for a network address. For hosts, use 255.255.255.255. However, for the ip address command, use a network mask, and for the global command, use a network address for both PAT addresses and when specifying a pool of global addresses.
For the conduit and access-list commands, precede host addresses with the host parameter and without specifying a mask.
The following are examples of commands in which a mask can be specified:
ip address inside 10.1.1.1 255.255.255.0ip address outside 209.165.201.1 255.255.255.224nat (inside) 1 10.1.1.0 255.255.255.0global (outside) 1 209.165.201.2 netmask 255.255.255.224static (inside,outside) 209.165.201.3 10.1.1.3 netmask 255.255.255.255access-list acl_out permit tcp any host 209.165.201.3 eq wwwaaa authentication include http outside 209.165.201.3 255.255.255.255 0 0 TACACS+route outside 0 0 209.165.201.4 1telnet 10.1.1.2 255.255.255.255In these examples, the ip address commands specify addresses for the inside and outside network interfaces. The ip address command only uses network masks. The inside interface is a Class A address, but only the last octet is used in the example network and therefore has a Class C mask. The outside interface is part of a subnet so the mask reflects the .224 subnet value.
The nat command lets users start connections from the inside network. Because a network address is specified, the class mask specified by the ip address inside command is used.
The global command provides a PAT (Port Address Translation) address to handle the translated connections from the inside. The global address is also part of the subnet and contains the same mask specified in the ip address outside command.
The static command maps an inside host to a global address for access by outside users. Host masks are always specified as 255.255.255.255.
The access-list command permits any outside host to access the global address specified by the static command. The host parameter is the same as if you specified 209.165.201.3 255.255.255.255.
The aaa command indicates that any users wishing to access the global address must be authenticated. Because authentication only occurs when users access the specified global which is mapped to a host, the mask is for a host. The "0 0" entry indicates any host and its respective mask.
The route statement specifies the address of the default router. The "0 0" entry indicates any host and its respective mask.
The telnet command specifies a host that can access the PIX Firewall unit's console using Telnet. Because it is a single host, a host mask is used.
If you are using subnet masks, refer to "Using Subnet Masks," to be sure that each IP address you choose for global or static addresses is in the correct subnet.
The subnet masks are also identified by the number of bits in the mask. Table D-3 lists subnet masks by the number of bits in the network ID.
The .255 mask indicates a single host in a network.
Uses for Subnet Information
Use subnet information to ensure that your host addresses are in the same subnet and that you are not accidentally using a network or broadcast address for a host.
The network address provides a way to reference all the addresses in a subnet, which you can use in the global, outbound, and static commands. For example, you can use the following net static statement to map global addresses 192.168.1.65 through 192.168.1.126 to local addresses 192.168.2.65 through 192.168.2.126: static (dmz1,dmz2) 192.168.1.64 192.168.2.64 netmask 255.255.255.192.
Subnet mask information is especially valuable when you have disabled Network Address Translation (NAT) using the nat 0 command. PIX Firewall requires that IP addresses on each interface be in different subnets.
However all the hosts on a PIX Firewall interface between the PIX Firewall and the router must be in the same subnet as well. For example, if you have an address such as 192.168.17.0 and you are not using NAT, you could use the 255.255.255.192 subnet mask for all three interfaces and use addresses 192.168.17.1 through 192.168.17.62 for the outside interface, 192.168.17.65 through 192.168.17.126 for the perimeter interface, and 192.168.17.129 through 192.168.17.190 for the inside interface.
With Limited IP Addresses
Another use for subnet mask information is for network planning when an Internet service provider (ISP) gives you a limited number of IP addresses and requires you to use a specific subnet mask. Use the information in this appendix to ensure that the outside addresses you choose are in the subnet for the appropriate subnet mask.
For example, if your ISP assigns you 192.168.17.176 with a subnet mask of .240, you can see in Table D-7, Subnet Number 12 for the .240 mask, that hosts can have IP addresses of 192.168.17.177 through 192.168.17.190. Because this only yields 14 hosts, you will probably use one for your router, another for the outside interface of the PIX Firewall, one for a static for a web server, if you have it, one for a static for your mail server, and the remaining 10 for global addresses. One of these addresses should be a PAT (Port Address Translation) address so that you do not run out of global addresses.
Addresses in the .128 Mask
Table D-4 lists valid addresses for the .128 subnet mask. This mask permits up to 2 subnets with enough host addresses for 126 hosts per subnet.
Table D-4 .128 Network Mask Addresses
Subnet Number Network Address Starting Host Address Ending Host Address Broadcast Address1
.0
.1
.126
.127
2
.128
.129
.254
.255
Addresses in the .192 Mask
Table D-5 lists valid addresses for the .192 subnet mask. This mask permits up to 4 subnets with enough host addresses for 62 hosts per subnet.
Addresses in the .224 Mask
Table D-6 lists valid addresses for the .224 subnet mask. This mask permits up to 8 subnets with enough host addresses for 30 hosts per subnet.
Addresses in the .240 Mask
Table D-7 lists valid addresses for the .240 subnet mask. This mask permits up to 16 subnets with enough host addresses for 14 hosts per subnet.
Addresses in the .248 Mask
Table D-8 lists valid addresses for the .248 subnet mask. This mask permits up to 32 subnets with enough host addresses for 6 hosts per subnet.
Addresses in the .252 Mask
Table D-9 lists valid addresses for the .252 subnet mask. This mask permits up to 64 subnets with enough host addresses for 2 hosts per subnet.
