A Novell IPX address is made up of a network number and a node element.
|
In text format, these elements are separated by a period, so IPX addresses are represented in the form ^Rnetwork.node^r.
|
In IPX packets the address is represented in a sequence of 80 bits.
|
The network number identifies a physical network.
|
The network number contains 32 of the 80 bits that make up the IPX address.
|
A network must have a number that is unique throughout the entire IPX internetwork.
|
A network administrator assigns the network number.
|
A network number can be expressed as eight hexadecimals.
|
For example, 0000004A is a valid network number.
|
When you use Cisco IOS software you don't have to enter all of the eight hexadecimal digits in a network number.
|
You can omit leading zeros.
|
So 0000004A becomes 4A.
|
The remaining 48 bits in the IPX address are used for the node number.
|
The node number is represented by a dotted triplet of four-digit hexadecimal numbers.
|
For example, an IPX network address might be 4A.0000.0C00.23FE.
|
Here 0000.0C00.23FE is the node number.
|
The node number contains the Media Access Control (MAC) address of the interface.
|
A MAC address is a standardized, link layer address that is required for every port or device connected to a LAN.

MAC addresses are 6 bytes long.
|
IPX's use of a MAC address for the node number allows sending nodes to predict what MAC address to use on a data link.
|
In contrast, because the host portion of an IP network address has no correlation to the MAC address, IP nodes must use the Address Resolution Protocol (ARP) to determine the destination MAC address.
|
The Address Resolution Protocol (ARP) is an Internet protocol used to map an IP address to a MAC address.

It is defined in RFC 826.
|
Since with IPX, the interface MAC address is part of the logical address, an Address Resolution Protocol (ARP) is not needed.
|
The MAC address is usually used to identify the interface.
@@
Encapsulation is the process of packaging upper layer protocol information and data into a frame.
|
A frame is an information unit whose source and destination is a link-layer entity.
|
Encapsulation uses the frame formats of the MAC protocols.
|
Many MAC protocols are from the IEEE 802.x series and specify particular header types which are used in IPX encapsulation.
|
Depending on what interface type is used, there are many different encapsulation formats available when using IPX routing.
|
In particular, Ethernet, Token Ring, FDDI, and PPP frame formats are used by IPX.
@@
<f>The most common IPX Ethernet packet encapsulation formats are Ethernet version 2 and Ethernet 802.3.
|
Ethernet version 2, also known as Ethernet II, is recommended for networks that handle both TCP/IP and IPX traffic.
|
It includes the standard Version 2 header, Destination and Source Address fields, followed by an Ether Type # field.
|
Ethernet 802.3 is Ethernet's ^Rraw^r format.
|
This means that an IEEE 802.3 header is used alone, without the usual IEEE 802.2 frame information.
|
802.3 is the default frame type for NetWare 3.1 but you cannot send IPX packets containing checksums encapsulated in 802.3 frames.
|
The default frame type for NetWare 4 and later releases is Ethernet 802.2.
|
For each Novell packet format, Cisco has a keyword.
|
For example, the keyword for Ethernet 2.0 is ^RArpa^r.
|
And the keyword for Ethernet 802.3 is ^RNovell Ether^r.
|
Novell developments now allow the encapsulation of IPX packets in standard 802.3 frames using Service Access Point (SAP).
|
The Service Access Point is a field defined by the 802.2 specification which is part of an address specification.

A SAP is also a logical interface between two adjacent OSI protocol layers.
|
The Cisco keyword for this frame format is ^RSAP^r.
|
The SubNetwork Access Protocol (SNAP) which extends the IEEE 802.2 headers has also been developed.
|
In Cisco terms this frame format is known as ^RSNAP^r.
|
Both SAP and SNAP encapsulations include the 802.2 Logical Link Control protocol (LLC).
|
This protocol handles error control, flow control, framing, and MAC sublayer addressing.
|
Cisco routers support all these frame encapsulation formats.
@@
Comparable raw (802.3), SAP, and SNAP frame formats used by other media access control protocols are supported by IPX.
|
FDDI has a raw 802.3 frame format which has the Cisco keyword ^RNovell-FDDI^r.
|
The FDDI SAP format consists of a standard FDDI MAC header followed by an 802.2 LLC header.
|
The FDDI SNAP format consists of an FDDI MAC header followed by an 802.2 SNAP LLC header.
|
The SAP and SNAP Cisco keywords remain the same.
|
Token Ring has no raw 802.3 format but has SAP and SNAP formats.
|
The token ring SAP format is the standard 802.5 MAC header followed by an 802.2 LLC header.
|
The SNAP format consists of the 802.5 header followed by a SNAP LLC header.
|
On serial interfaces Novell IPX uses PPP's HDLC encapsulation.
|
The PPP/HDLC frame format has the following fields:

 Flag - marks start or end of frame
|
 Address - broadcast address
|
 Control - similar to LLC
|
 Protocol - encapsulated protocol 
  name
|
 Datagram - datagram contained
|
 Frame check sequence - error 
  control
@@
The IPX packet is similar to an XNS packet and consists of two parts:

 A 30-byte (minimum) IPX
  header - containing 
  various packet fields
|
 Data - including the 
  header of a higher level
  protocol
|
The IPX Packet format has several fields.
|
The "checksum" field is a 16-bit field that is set to 1's.
|
The checksum is set to all 1's (FFFF) since it is not used.
|
The packet length field is a 16-bit field.
|
It specifies the length, in bytes, of the complete IPX datagram.
|
The packet length is at least 30 bytes.
|
IPX packets can be any size up to the particular media maximum transmission unit (MTU).
|
The MTU is the maximum packet size in bytes that a particular interface can handle.
|
For example, Ethernet 2.0 packets are limited to a packet size of 1500 bytes.
|
The maximum packet size that IPX can transmit is 65535 bytes.
@@
The transport control field is 8 bits long.
|
The transport control field indicates the number of routers that the packet has passed through.
|
On a RIP-based IPX router, IPX packets whose transport control field reaches 16 are discarded under the assumption that a routing loop might be occurring.
|
Sending nodes always set this field to zero when building an IPX packet.
|
Routers receiving packets that require more routing increment the field by one and then route the packet.
|
The packet type field is also an 8-bit field.
|
The information in this field specifies which upper layer protocol, such as NCP, SAP, SPX, NetBIOS, or RIP, is to receive the packet's information.
|
When the value is 5, the protocol will be SPX.
|
When the value is 17, NCP is specified.
|
The destination network, destination node, and destination socket fields specify relevant destination information.
|
The destination network is the number of the network to which the destination node is attached.
|
When this field is zero, the sending and receiving nodes are assumed to be on the same network segment.
|
The destination node field represents the physical address of the destination node.
|
The socket field is the socket address of the packet destination process.
|
A socket is a software structure operating as a communications end point within a network device.
|
The corresponding source fields specify the following source information:

 source network
|
 source node
|
 source socket
|
The upper layer data field contains information required for upper layer processes such as headers of upper level protocols like NCP or SPX.
|
Data for these headers is contained in the data portion of the IPX packet.
@@
When configuring IPX on Cisco routers you need to plan your network address assignment.
|
During the course of IPX configuration you assign network numbers to individual interfaces.
|
This enables IPX routing on those interfaces.
|
Your first step in configuring IPX is to enable IPX routing.
|
This involves entering the IPX Routing command in Configuration mode, and then entering the node number of the router.
|
If you do not specify the node number, the Cisco IOS software uses the hardware Media Access Control (MAC) address currently assigned to it.
|
This might be, for example, the MAC address of the first Ethernet, Token Ring or FDDI interface card.
|
This is the format of the command to enable IPX routing.
|
The IPX host address is defaulted to the first IEEE conformance interface, Ethernet O.
|
Once you have enabled IPX routing on an interface, you can specify an encapsulation or frame type for packets being transmitted.
|
A single interface can support either a single network or multiple logical networks.
|
If you use a single network, you can configure any encapsulation type as long as it matches that of the clients and servers using that network number.
|
This is the format of the command used to assign a network number to an interface that supports a single network.
|
After you have entered ^Ripx network^r you add in the network number you have specified, for example, ^R4A^r.
|
You then have the option of entering an encapsulation type using the appropriate Novell keyword.
|
For example you may enter ^Rencapsulation arpa^r for an Ethernet interface.
@@
Unlike single networks, if you are assigning network numbers to an interface that supports multiple networks you must specify different encapsulation types for each network.
|
Multiple networks assigned to an interface all share the same physical medium.
|
Assigning different encapsulation types allows Cisco IOS software to identify the packets that belong to each network.
|
A network administrator is configuring IPX using a single Ethernet cable.
|
There are four packet encapsulation types supported for Ethernet.
|
Up to four IPX networks can be configured in accordance with the four Ethernet encapsulation types available.
|
As with single networks, the encapsulation type for each of the multiple networks should match the servers and clients using the same network number.
|
To assign network numbers to interfaces that support multiple networks, you usually use subinterfaces.
|
Using subinterfaces allows several logical interfaces or networks to be associated with a single hardware or physical interface.
|
To enable IPX on the multiple networks you first enable IPX routing and then follow two other configuration steps.
|
For more information on enabling IPX routing and practice sessions, see the CBT Systems course ^RImplementing IPX RIP and NLSP on Cisco routers^r.
|
Here we see the command that allows you to complete the first step of specifying a subinterface.
|
You enter ^Rinterface^r, then the interface type, for example Ethernet.
|
The ^Bnumber^b segment refers to a port, connector, or interface card number.
|
Following a period you type a subinterface number that is associated with the interface referred to earlier in the command.
|
A subinterface number can be in the range 1 to 4294967293.
|
Once you have specified the subinterface, the second step is identical to that used to enable IPX on a single network.
|
You enter ^Ripx network^r, add in the appropriate network number and assign an encapsulation type.
|
To configure more than one subinterface, you simply repeat these two steps, remembering to vary the encapsulation type each time.
|
Another possible way of configuring multiple networks on an interface is to first assign a primary and then a secondary network.

However, primary and secondary networks will not be supported by future Cisco IOS software releases.
|
When you specify an individual subinterface, any interface configuration parameters you use, for example a routing update timer, are applied only to that subinterface.
