Internetwork Packet Exchange IPX

IPX is Novell's layer 3 protocol. The addressing method used in IPX is a network with the source MAC address appended; 32 bits is used to identify the IPX network; and 48 bits identifies the host portion, because the burned-in address (BIA) is used to identify the node or client address to IPX. IPX allows clients and servers to communicate locally or remotely. The routing protocol can be IPX RIP or Cisco propriety EIGRP on Cisco routers. Figure 2.8 shows the relationship between the OSI model and IPX by comparing them side by side.

Figure 2.8 shows were each protocol fits into the IPX model and the corresponding relationship to the OSI model. For example, IPX is routed using the IPX RIP routing protocol that provides connectionless routing services at the Network layer. IPX can also be routed using other routing protocols, which will be discussed in Chapter 7. The Sequenced Packet Exchange (SPX) protocol provides connection-orientated services to IPX clients. Comparing IPX to IP, you can see that the routing protocol at layer 3 for IP, such as OSPF, compares with the routing protocol for IPX, called IPX RIP. Both protocols service different applications. TCP (layer 4 of the TCP/IP model) compares to SPX (layer 4 of IPX

Application	NetWare Control Program (NCP)	Service Advertising Protocol (SAP)	Routing Information Protocol (IPX RIP)
Presentation
Session	NetBIOS
Transport	SPX
Network	IPX
Data Link	Ethernet and token ring are examples UTP, coax, and fiber are examples
Physical

Figure 2.8 IPX model compared to the OSI model.

Figure 2.8 IPX model compared to the OSI model.

model). Let's examine the frame format for IPX, which is outlined in Figure 2.9. The following summarizes each field:

► Checksum—Specifies a setting of FFFF (not used).

► Packet Length—Specifies the length of the IPX packet.

► Transport Control—Indicates the hop count (initially, it is set to 0).

► Packet Type—Indicates the payload type. Examples include 1 for RIP, 4 for Services Advertisement Protocol (SAP), and 5 for Sequenced Packet Exchange (SPX).

Bits Specified
	Checksum (16 bits)
	Packet Length (16 bits)
	Transport Control (8 bits)	Packet Type (8 bits)
	Destination Network (32 bits)
	Destination Node (48 bits)
	Destination Socket (16 bits)
	Source Network (32 bits)
	Source Node (48 bits)
	Source Socket (16 bits)
	Data (Variable)

Figure 2.9 IPX frame format.

Figure 2.9 IPX frame format.

► Destination Network—Specifies the 32-bit destination network.

► Destination Node—Identifies the MAC address of the destination.

► Destination Socket—Serves as a 16-bit indicator that describes what function is being used. Examples include 0451 for the Application layer's protocol called NCP (NetWare Control Program). Other examples include 0452 for SAP and 0453 for RIP information.

► Source Network—Specifies a 32-bit source network.

► Source Node—Specifies a 48-bit source node.

► Source Socket—Indicates that the source socket is not using a well-known number. The source socket number is a random number in most cases.

► Data—Follows the header. There is no checksum made on the packet.

The Cisco router supports a number of encapsulation types for IPX networks.

Novell and Cisco use different names for each. On a Cisco router, when you configure IPX, you can also specify the encapsulation type, as shown in the following IOS display:

R1(config-if)#ipx encapsulation ?

arpa IPX Ethernet_II

hdlc HDLC on serial links novell-ether IPX Ethernet_802.3

sap IEEE 802.2 on Ethernet, Token Ring, and FDDI

snap IEEE 802.2 SNAP on Ethernet, Token Ring, and FDDI

Table 2.3 shows the available options when configuring an interface for IPX. Be aware that the encapsulation on end systems must match encapsulation on the local segment; otherwise, there will be no communication between the relevant devices.

All the encapsulations shown in Table 2.3 have different frame formats. The frame formats are as follows, using Novell names (the abbreviations are spelled out in the subsequent bulleted list):

Table 2.3 Encapsulation support on Cisco routers.
Novell Name	Cisco Name
Ethernet_II	ARPA
Ethernet_802.3	Novell-Ether
Ethernet_802.2	SAP
Ethernet_SNAP	SNAP

► Ethemet_II—DA\SA\Type Field|Data|FCS

► Ethernet802.3—DA\SA\Length\Data\FCS

► Ethernet802.2—DA\SA\Length\DSAP\SSAP\Control\Data\FCS

► Ethernet_SNAP—DA\SA\Length\DSAP\SSAP\Control\OUI\Type\ Data\FCS

In IPX over Token Ring (which is covered in Chapter 4), the frame format is as follows:

SD\AC\FC\DA\SA\RIF\DSAP\SSAP\Control\Data\ED\FCS\FS The abbreviations used in the preceding Ethernet formats are defined as:

► DA—Destination address.

► Type Field—A field that tells higher layers what protocol is carried.

► DSAP—Destination service access point; set to E0.

► SSAP—Source service access point; set to E0.

► Control—Set to 03 for connectionless services. IPX is a connectionless protocol.

► SNAP Header (OUI and Type field)—The unique organization identifier (OUI).

► FCS—Frame control sequence used for error control.

IP Vs. IPX

It is important to differentiate between IP and IPX. IP and IPX are both connectionless protocols and are functions provided at layer 3 of the OSI model. The differences between IP and IPX are primarily the addressing scheme and frame formats. IP addresses are represented with 32 bits, and IPX addresses are 48 bits in length.

Continue reading here: Transmission Control Protocol TCP

Was this article helpful?

0 0