Novell IPX Configuration Guide, Cisco IOS Release 15SY

Configuring Novell IPX

Note Effective with Cisco IOS Release 15.1(3)S, 15.2(2)T, 15.1(1)SY, and 15.2(1)E this feature is not supported in Cisco IOS software.

Feature History

This chapter describes how to configure Novell Internetwork Packet Exchange (IPX) and provides configuration examples. For a complete description of the IPX commands in this chapter, refer to the Cisco IOS Novell IPX Command Reference publication. To locate documentation of other commands that appear in this chapter, use the command reference master index or search online.

For the latest feature information and caveats, see the release notes for your platform and software release. Additionally, use Cisco Feature Navigator to find information about feature, platform, and software image support. To access Cisco Feature Navigator, go to http://www.cisco.com/go/cfn.

IPX Addresses

An IPX network address consists of a network number and a node number expressed in the format network.node.

Network Numbers

The network number identifies a physical network. It is a 4-byte (32-bit) quantity that must be unique throughout the entire IPX internetwork. The network number is expressed as hexadecimal digits. The maximum number of digits allowed is eight.

The Cisco IOS software does not require that you enter all eight digits; you can omit leading zeros.

Node Numbers

The node number identifies a node on the network. It is a 48-bit quantity, represented by dotted triplets of four-digit hexadecimal numbers.

If you do not specify a node number for a router to be used on WAN links, the Cisco IOS software uses the hardware MAC address currently assigned to it as its node address. This is the MAC address of the first Ethernet, Token Ring, or FDDI interface card. If there are no valid IEEE interfaces, the Cisco IOS software randomly assigns a node number using a number that is based on the system clock.

IPX Address Example

The following example shows how to configure an IPX network address:

4a.0000.0c00.23fe

In this example, the network number is 4a (more specifically, it is 0000004a), and the node number is 0000.0c00.23fe. All digits in the address are hexadecimal.

IPX Configuration Task List

To configure IPX routing, perform the tasks in the following sections:

•Configuring IPX Routing (Required)

•Configuring IPX Enhanced IGRP (Optional)

•Configuring IPX and SPX over WANs (Optional)

•Controlling Access to IPX Networks (Optional)

•Tuning IPX Network Performance (Optional)

•Shutting Down an IPX Network (Optional)

•Configuring IPX Accounting (Optional)

•Configuring IPX Between LANs (Optional)

•Configuring IPX Between VLANs (Optional)

•Configuring IPX Multilayer Switching (Optional)

•Monitoring and Maintaining the IPX Network (Optional)

See the "Novell IPX Configuration Examples" section at the end of this chapter for configuration examples.

Configuring IPX Routing

You configure IPX routing by first enabling it on the router and then configuring it on each interface.

Optionally, you can route IPX on some interfaces and transparently bridge it on other interfaces. You can also route IPX traffic between routed interfaces and bridge groups, or route IPX traffic between bridge groups.

To configure IPX routing, perform the tasks in the following sections. The first two tasks are required; the rest are optional.

•Enabling IPX Routing (Required)

•Assigning Network Numbers to Individual Interfaces (Required)

•Enabling Concurrent Routing and Bridging (Optional)

•Configuring Integrated Routing and Bridging (Optional)

IPX Default Routes

In IPX, a default route is the network where all packets for which the route to the destination address is unknown are forwarded.

Original Routing Information Protocol (RIP) implementations allowed the use of network -2 (0xFFFFFFFE) as a regular network number in a network. With the inception of NetWare Link Services Protocol (NLSP), network -2 is reserved as the default route for NLSP and RIP. Both NLSP and RIP routers should treat network -2 as a default route. Therefore, you should implement network -2 as the default route regardless of whether you configure NLSP in your IPX network.

By default, Cisco IOS software treats network -2 as the default route. You should ensure that your IPX network does not use network -2 as a regular network. If, for some reason, you must use network -2 as a regular network, you can disable the default behavior. To do so, see the "Adjusting Default Routes" section later in this chapter.

For more background information on how to handle IPX default routes, refer to the Novell NetWare Link Services Protocol (NLSP) Specification, Revision 1.1 publication.

Enabling IPX Routing

The first step in enabling IPX routing is to enable it on the router. If you do not specify the node number of the router to be used on WAN links, the Cisco IOS software uses the hardware MAC address currently assigned to it as its node address. This is the MAC address of the first Ethernet, Token Ring, or FDDI interface card. If there are no valid IEEE interfaces, the Cisco IOS software randomly assigns a node number using a number that is based on the system clock.

To enable IPX routing, use the following command in global configuration mode:

For an example of how to enable IPX routing, see the "IPX Routing Examples" section at the end of this chapter.

Caution If you plan to use DECnet and IPX routing concurrently on the same interface, you should enable DECnet routing first, then enable IPX routing without specifying the optional MAC node number. If you enable IPX before enabling DECnet routing, routing for IPX will be disrupted because DECnet forces a change in the MAC-level node number.

Assigning Network Numbers to Individual Interfaces

After you have enabled IPX routing, you enable IPX routing on the individual interfaces by assigning network numbers to those interfaces.

You enable IPX routing on interfaces that support a single network or multiple networks.

When you enable IPX routing on an interface, you can also specify an encapsulation (frame type) to use for packets being sent on that network. Table 1 lists the encapsulation types you can use on IEEE interfaces and shows the correspondence between Cisco naming conventions and Novell naming conventions for the encapsulation types.

Note The SNAP encapsulation type is not supported and should not be configured on any IPX interfaces that are attached to a FDDI-Ethernet bridge.

Assigning Network Numbers to Individual Interfaces Task List

The following sections describe how to enable IPX routing on interfaces that support a single network and on those that support multiple networks. To enable IPX routing on an interface, you must perform one of the tasks:

•Assigning Network Numbers to Interfaces That Support a Single Network (Required)

•Assigning Network Numbers to Interfaces That Support Multiple Networks (Required)

•Setting the Encapsulation Type for Subinterfaces (Required)

Assigning Network Numbers to Interfaces That Support a Single Network

A single interface can support a single network or multiple logical networks. For a single network, you can configure any encapsulation type. Of course, it should match the encapsulation type of the servers and clients using that network number.

To assign a network number to an interface that supports a single network, use the following command in interface configuration mode:

If you specify an encapsulation type, be sure to choose the one that matches the one used by the servers and clients on that network. Novell-ether or ARPA encapsulations cannot be used for FDDI-Ethernet bridged IPX traffic. Use SAP encapsulations on originating and destination IPX interfaces that are attached to the FDDI-Ethernet bridge. See Table 1 for a list of encapsulation types you can use on IEEE interfaces.

For an example of how to enable IPX routing, see the "IPX Routing Examples" section at the end of this chapter.

Assigning Network Numbers to Interfaces That Support Multiple Networks

When assigning network numbers to an interface that supports multiple networks, you must specify a different encapsulation type for each network. Because multiple networks share the physical medium, the Cisco IOS software is allowed to identify the packets that belong to each network. For example, you can configure up to four IPX networks on a single Ethernet cable, because four encapsulation types are supported for Ethernet. Remember, the encapsulation type should match the servers and clients using the same network number. See Table 1 for a list of encapsulation types you can use on IEEE interfaces.

There are two ways to assign network numbers to interfaces that support multiple networks. You can use subinterfaces or primary and secondary networks.

Setting the Encapsulation Type for Subinterfaces

You typically use subinterfaces to assign network numbers to interfaces that support multiple networks.

A subinterface is a mechanism that allows a single physical interface to support multiple logical interfaces or networks. That is, several logical interfaces or networks can be associated with a single hardware interface. Each subinterface must use a distinct encapsulation, and the encapsulation must match that of the clients and servers using the same network number.

Note When enabling NLSP and configuring multiple encapsulations on the same physical LAN interface, you must use subinterfaces. You cannot use secondary networks.

Any interface configuration parameters that you specify on an individual subinterface are applied to that subinterface only.

To configure multiple IPX networks on a physical interface using subinterfaces, use the following commands beginning in global configuration mode:

Note You cannot configure more than 200 IPX interfaces on a router using the ipx network command.

To configure more than one subinterface, repeat these two steps. See Table 1 for a list of encapsulation types you can use on IEEE interfaces.

For examples of configuring multiple IPX networks on an interface, see the "IPX Routing on Multiple Networks Examples" section at the end of this chapter.

Primary and Secondary Networks

When assigning network numbers to interfaces that support multiple networks, you can also configure primary and secondary networks.

The first logical network you configure on an interface is considered the primary network. Any additional networks are considered secondary networks. Again, each network on an interface must use a distinct encapsulation and it should match that of the clients and servers using the same network number.

Any interface configuration parameters that you specify on this interface are applied to all the logical networks. For example, if you set the routing update timer to 120 seconds, this value is used on all four networks.

To use primary and secondary networks to configure multiple IPX networks on an interface, use the following commands in interface configuration mode:

To configure more than one secondary network, repeat these steps as appropriate. See Table 1 for a list of encapsulation types you can use on IEEE interfaces.

Note When enabling NLSP and configuring multiple encapsulations on the same physical LAN interface, you must use subinterfaces. You cannot use secondary networks.

Enabling Concurrent Routing and Bridging

You can route IPX on some interfaces and transparently bridge it on other interfaces simultaneously. To enable this type of routing, you must enable concurrent routing and bridging. To enable concurrent routing and bridging, use the following command in global configuration mode:

Configuring Integrated Routing and Bridging

Integrated routing and bridging (IRB) enables a user to route IPX traffic between routed interfaces and bridge groups, or route IPX traffic between bridge groups. Specifically, local or unroutable traffic is bridged among the bridged interfaces in the same bridge group. Routable traffic is routed to other routed interfaces or bridge groups. Using IRB, you can do the following:

•Switch packets from a bridged interface to a routed interface

•Switch packets from a routed interface to a bridged interface

•Switch packets within the same bridge group

For more information about configuring integrated routing and bridging, refer to the "Configuring Transparent Bridging" chapter in the Cisco IOS Bridging and IBM Networking Configuration Guide.

Configuring IPX Enhanced IGRP

Enhanced IGRP is an enhanced version of the Interior Gateway Routing Protocol (IGRP) developed by Cisco. Enhanced IGRP uses the same distance vector algorithm and distance information as IGRP. However, the convergence properties and the operating efficiency of Enhanced IGRP have improved significantly over IGRP.

The convergence technology is based on research conducted at SRI International and employs an algorithm referred to as the Diffusing Update Algorithm (DUAL). This algorithm guarantees loop-free operation at every instant throughout a route computation, and allows all routers involved in a topology change to synchronize at the same time. Routers that are not affected by topology changes are not involved in recomputations. The convergence time with DUAL rivals that of any other existing routing protocol.

Enhanced IGRP Features

Enhanced IGRP offers the following features:

•Fast convergence—The DUAL algorithm allows routing information to converge as quickly as any currently available routing protocol.

•Partial updates—Enhanced IGRP sends incremental updates when the state of a destination changes, instead of sending the entire contents of the routing table. This feature minimizes the bandwidth required for Enhanced IGRP packets.

•Less CPU usage than IGRP—Full update packets need not be processed each time they are received.

•Neighbor discovery mechanism—This feature is a simple hello mechanism used to learn about neighboring routers. It is protocol-independent.

•Scaling—Enhanced IGRP scales to large networks.

Enhanced IGRP Components

Enhanced IGRP has four basic components discussed in the following sections:

•Neighbor Discovery/Recovery

•Reliable Transport Protocol

•DUAL Finite-State Machine

•Protocol-Dependent Modules

Neighbor Discovery/Recovery

Neighbor discovery/recovery is the process that routers use to dynamically learn of other routers on their directly attached networks. Routers must also discover when their neighbors become unreachable or inoperative. The router achieves neighbor discovery/recovery with low overhead by periodically sending small hello packets. As long as hello packets are received, a router can determine that a neighbor is alive and functioning. Once this status is determined, the neighboring devices can exchange routing information.

Reliable Transport Protocol

The reliable transport protocol is responsible for guaranteed, ordered delivery of Enhanced IGRP packets to all neighbors. It supports intermixed transmission of multicast and unicast packets. Some Enhanced IGRP packets must be sent reliably, and others need not be. For efficiency, reliability is provided only when necessary. For example, on a multiaccess network that has multicast capabilities (such as Ethernet) it is not necessary to send hellos reliably to all neighbors individually. Therefore, Enhanced IGRP sends a single multicast hello with an indication in the packet informing the receivers that the packet need not be acknowledged. Other types of packets (such as updates) require acknowledgment, which is indicated in the packet. The reliable transport has a provision to send multicast packets quickly when there are unacknowledged packets pending. This provision helps ensure that convergence time remains low in the presence of varying speed links.

DUAL Finite-State Machine

The DUAL finite-state machine embodies the decision process for all route computations. It tracks all routes advertised by all neighbors. DUAL uses the distance information (known as a metric) to select efficient, loop-free paths. DUAL selects routes to be inserted into a routing table based on feasible successors. A successor is a neighboring router used for packet forwarding that has a least-cost path to a destination that is guaranteed not to be part of a routing loop. When there are no feasible successors but there are neighbors advertising the destination, a recomputation must occur. This is the process whereby a new successor is determined. The amount of time it takes to recompute the route affects the convergence time. Recomputation is processor-intensive. It is advantageous to avoid recomputation if it is not necessary. When a topology change occurs, DUAL will test for feasible successors. If there are feasible successors, it will use any it finds in order to avoid unnecessary recomputation.

Protocol-Dependent Modules

The protocol-dependent modules are responsible for network layer protocol-specific tasks. They are also responsible for parsing Enhanced IGRP packets and informing DUAL of the new information received. Enhanced IGRP asks DUAL to make routing decisions, but the results are stored in the IPX routing table. Also, Enhanced IGRP is responsible for redistributing routes learned by other IPX routing protocols.

IPX Enhanced IGRP Configuration Task List

To enable IPX Enhanced IGRP, perform the tasks in the following sections. Only the first task is required; the remaining tasks are optional.

•Enabling IPX Enhanced IGRP (Required)

•Customizing Link Characteristics (Optional)

•Customizing the Exchange of Routing and Service Information (Optional)

•Querying the Backup Server (Optional)

Enabling IPX Enhanced IGRP

To create an IPX Enhanced IGRP routing process, use the following commands beginning in global configuration mode:

To associate multiple networks with an Enhanced IGRP routing process, you can repeat the preceding two steps.

For an example of how to enable Enhanced IGRP, see the "IPX Enhanced IGRP Example" section at the end of this chapter.

Customizing Link Characteristics

You might want to customize the Enhanced IGRP link characteristics. The following sections describe these customization tasks:

•Configuring the Percentage of Link Bandwidth Used by Enhanced IGRP (Optional)

•Configuring Maximum Hop Count (Optional)

•Adjusting the Interval Between Hello Packets and the Hold Time (Optional)

Configuring the Percentage of Link Bandwidth Used by Enhanced IGRP

By default, Enhanced IGRP packets consume a maximum of 50 percent of the link bandwidth, as configured with the bandwidth interface subcommand. If a different value is desired, use the ipx bandwidth-percent command. This command may be useful if a different level of link utilization is required, or if the configured bandwidth does not match the actual link bandwidth (it may have been configured to influence route metric calculations).

To configure the percentage of bandwidth that may be used by Enhanced IGRP on an interface, use the following command in interface configuration mode:

For an example of how to configure the percentage of Enhanced IGRP bandwidth, see the "IPX Enhanced IGRP Bandwidth Configuration Example" section at the end of this chapter.

Configuring Maximum Hop Count

Note Although adjusting the maximum hop count is possible, it is not recommended for Enhanced IGRP. We recommend that you use the default value for the maximum hop count of Enhanced IGRP.

By default, IPX packets whose hop count exceeds 15 are discarded. In larger internetworks, this maximum hop count may be insufficient. You can increase the hop count to a maximum of 254 hops for Enhanced IGRP. To modify the maximum hop count, use the following command in global configuration mode:

Adjusting the Interval Between Hello Packets and the Hold Time

You can adjust the interval between hello packets and the hold time.

Routers periodically send hello packets to each other to dynamically learn of other devices on their directly attached networks. Routers use this information to discover their neighbors and to discover when their neighbors become unreachable or inoperative.

By default, hello packets are sent every 5 seconds. The exception is on low-speed, nonbroadcast multiaccess (NBMA) media, where the default hello interval is 60 seconds. Low speed is considered to be a rate of T1 or slower, as specified with the bandwidth interface configuration command. The default hello interval remains 5 seconds for high-speed NBMA networks.

Note For the purposes of Enhanced IGRP, Frame Relay and SMDS networks may or may not be considered to be NBMA. These networks are considered NBMA if the interface has not been configured to use physical multicasting; otherwise they are considered not to be NBMA.

You can configure the hold time on a specified interface for a particular Enhanced IGRP routing process designated by the autonomous system number. The hold time is advertised in hello packets and indicates to neighbors the length of time they should consider the sender valid. The default hold time is three times the hello interval, or 15 seconds.

On very congested and large networks, 15 seconds may not be sufficient time for all routers to receive hello packets from their neighbors. In this case, you may want to increase the hold time. To increase the hold time, use the following command in interface configuration mode:

To change the interval between hello packets, use the following command in interface configuration mode:

Note Do not adjust the hold time without consulting with Cisco technical support.

Customizing the Exchange of Routing and Service Information

You might want to customize the exchange of routing and service information. The following sections describe these customization tasks:

•Redistributing Routing Information (Optional)

•Disabling Split Horizon (Optional)

•Controlling the Advertising of Routes in Routing Updates (Optional)

•Controlling the Processing of Routing Updates (Optional)

•Controlling SAP Updates (Optional)

•Controlling the Advertising of Services in SAP Updates (Optional)

•Controlling the Processing of SAP Updates (Optional)

Redistributing Routing Information

By default, the Cisco IOS software redistributes IPX RIP routes into Enhanced IGRP, and vice versa.

To disable route redistribution, use the following command in IPX-router configuration mode:

The Cisco IOS software does not automatically redistribute NLSP routes into Enhanced IGRP routes and vice versa. To configure this type of redistribution, use the following commands beginning in global configuration mode:

For an example of how to enable redistribution of Enhanced IGRP and NLSP, see the "Enhanced IGRP and NLSP Route Redistribution Example" section at the end of this chapter.

Disabling Split Horizon

Split horizon controls the sending of Enhanced IGRP update and query packets. If split horizon is enabled on an interface, these packets are not sent for destinations if this interface is the next hop to that destination.

By default, split horizon is enabled on all interfaces.

Split horizon blocks information about routes from being advertised by the Cisco IOS software out any interface from which that information originated. This behavior usually optimizes communication among multiple routers, particularly when links are broken. However, with nonbroadcast networks (such as Frame Relay and SMDS), situations can arise for which this behavior is less than ideal. For these situations, you can disable split horizon.

To disable split horizon, use the following command in interface configuration mode:

Note Split horizon cannot be disabled for RIP or SAP, only for Enhanced IGRP.

Controlling the Advertising of Routes in Routing Updates

To control which devices learn about routes, you can control the advertising of routes in routing updates. To control this advertising, use the following command in router configuration mode:

Controlling the Processing of Routing Updates

To control the processing of routes listed in incoming updates, use the following command in router configuration mode:

Controlling SAP Updates

If IPX Enhanced IGRP peers are found on an interface, you can configure the Cisco IOS software to send SAP updates either periodically or when a change occurs in the SAP table. When no IPX Enhanced IGRP peer is present on the interface, periodic SAPs are always sent.

On serial lines, by default, if an Enhanced IGRP neighbor is present, the Cisco IOS software sends SAP updates only when the SAP table changes. On Ethernet, Token Ring, and FDDI interfaces, by default, the software sends SAP updates periodically. To reduce the amount of bandwidth required to send SAP updates, you might want to disable the periodic sending of SAP updates on LAN interfaces. This feature should only be disabled when all nodes out of this interface are Enhanced IGRP peers; otherwise, loss of SAP information on the other nodes will result.

To send SAP updates only when a change occurs in the SAP table, use the following command in interface configuration mode:

To send SAP updates only when a change occurs in the SAP table and to send only the SAP changes, use the following command in interface configuration mode:

When you enable incremental SAP using the ipx sap-incremental eigrp rsup-only command, Cisco IOS software disables the exchange of route information via Enhanced IGRP for that interface.

To send periodic SAP updates, use the following command in interface configuration mode:

For an example of how to configure SAP updates, see the "Enhanced IGRP SAP Update Examples" section at the end of this chapter.

To disable split horizon for incremental SAP, use the following command in interface configuration mode:

Note IPX incremental SAP split horizon is off for WAN interfaces and subinterfaces, and on for LAN interfaces. The global default stays off. The interface setting takes precedence if the interface setting is modified or when both the global and interface settings are unmodified. The global setting is used only when the global setting is modified and the interface setting is unmodified.

Controlling the Advertising of Services in SAP Updates

To control which devices learn about services, you can control the advertising of these services in SAP updates. To control this advertising, use the following command in router configuration mode:

For a configuration example of controlling the advertisement of SAP updates, see the "Advertisement and Processing of SAP Update Examples" section at the end of this chapter.

Controlling the Processing of SAP Updates

To control the processing of routes listed in incoming updates, use the following command in router configuration mode:

For a configuration example of controlling the processing of SAP updates, see the "Advertisement and Processing of SAP Update Examples" section at the end of this chapter.

Querying the Backup Server

The backup server table is a table kept for each Enhanced IGRP peer. It lists the IPX servers that have been advertised by that peer. If a server is removed from the main server table at any time and for any reason, the Cisco IOS software examines the backup server table to learn if this just-removed server is known by any of the Enhanced IGRP peers. If it is, the information from that peer is advertised back into the main server table just as if that peer had readvertised the server information to this router. Using this method to allow the router to keep the backup server table consistent with what is advertised by each peer means that only changes to the table must be advertised between Enhanced IGRP routers; full periodic updates need not be sent.

By default, the Cisco IOS software queries its own copy of the backup server table of each Enhanced IGRP neighbor every 60 seconds. To change this interval, use the following command in global configuration mode:

Configuring IPX and SPX over WANs

You can configure IPX over dial-on-demand routing (DDR), Frame Relay, PPP, SMDS, and X.25 networks. For more information about dial-on-demand routing (DDR) refer to the Cisco IOS Dial Technologies Configuration Guide. For more information about Frame Relay, SMDS, and X.25 refer to the Cisco IOS Wide-Area Networking Configuration Guide.

When you configure IPX over PPP, address maps are not necessary for this protocol. Also, you can enable IPX header compression over point-to-point links to increase available useful bandwidth of the link and reduce response time for interactive uses of the link.

You can use fast-switching IPX serial interfaces configured for Frame Relay and SMDS, and you can use fast-switching Subnetwork Access Protocol (SNAP)-encapsulated packets over interfaces configured for ATM.

Additionally, you can configure the IPXWAN protocol.

For an example of how to configure IPX over a WAN interface, see the "IPX over a WAN Interface Example" section at the end of this chapter.

Configuring IPX over DDR

IPX sends periodic watchdog keepalive packets from servers to clients after a client session has been idle for approximately 5 minutes. On a DDR link, a call would be made every 5 minutes, regardless of whether there were data packets to send. You can prevent these calls from being made by configuring the Cisco IOS software to respond to the watchdog keepalive packets of a server on behalf of a remote client—sometimes referred to as spoofing the server. Spoofing makes a server view a client as always connected, even when it is not, thus reducing the number of available licenses. Users can set the duration of IPX watchdog spoofing and periodically disable it so that Novelle NetWare servers can clean up inactive connections.

When configuring IPX over DDR, you might want to disable the generation of these packets so that a call is not made every 5 minutes. A call made every 5 minutes is not an issue for the other WAN protocols, because they establish dedicated connections rather than establishing connections only as needed.

Use the ipx watchdog-spoof command to enable and set the duration of watchdog spoofing. You can specify the number of consecutive hours spoofing is to stay enabled and the number of minutes spoofing is to stay disabled. The server can clean up inactive connections when spoofing is disabled. Be sure that fast switching and autonomous switching are disabled on the serial interface before using this command.

To enable watchdog spoofing, use the following command in interface configuration mode:

To keep the serial interface idle when only watchdog packets are being sent, refer to the tasks described in the "Deciding and Preparing to Configure DDR" chapter of the Cisco IOS Dial Technologies Configuration Guide. For an example of configuring IPX over DDR, see the "IPX over DDR Example" section at the end of this chapter.

Configuring SPX Spoofing over DDR

Sequenced Packet Exchange (SPX) sends periodic keepalive packets between clients and servers. Similar to IPX watchdog packets, these are keepalive packets that are sent between servers and clients after the data has stopped being transferred. On pay-per-packet or byte networks, these packets can incur large customer telephone connection charges for idle time. You can prevent these calls from being made by configuring the Cisco IOS software to respond to the keepalive packets on behalf of a remote system.

When configuring SPX over DDR, you might want to disable the generation of these packets so that a call has the opportunity to go idle. Disabling the generation of packets may not be an issue for the other WAN protocols, because they establish dedicated connections rather than establishing connections only as needed.

To keep the serial interface idle when only keepalive packets are being sent, refer to the tasks described in the "Deciding and Preparing to Configure DDR" chapter of the Cisco IOS Dial Technologies Configuration Guide.

For an example of how to configure SPX spoofing over DDR, see the "IPX over DDR Example" section at the end of this chapter.

Configuring IPX Header Compression

You can configure IPX header compression over point-to-point links. With IPX header compression, a point-to-point link can compress IPX headers only, or the combined IPX and NetWare Core Protocol headers. Currently, point-to-point links must first negotiate IPX header compression via IPXCP or IXPWAN. The Cisco IOS software supports IPX header compression as defined by RFC 1553.

For details on configuring IPX header compression, refer to the "Configuring Medial-Independent PPP and Multilink PPP" chapter in the Cisco IOS Dial Technologies Configuration Guide.

Configuring the IPXWAN Protocol

The Cisco IOS software supports the IPXWAN protocol, as defined in RFC 1634. IPXWAN allows a router that is running IPX routing to connect via a serial link to another router, possibly from another manufacturer, that is also routing IPX and using IPXWAN.

IPXWAN is a connection startup protocol. Once a link has been established, IPXWAN incurs little or no overhead.

You can use the IPXWAN protocol over PPP. You can also use it over HDLC; however, the devices at both ends of the serial link must be Cisco routers.

To configure IPXWAN on a serial interface, use the following commands in interface configuration mode:

Controlling Access to IPX Networks

To control access to IPX networks, first create access lists and then apply them to individual interfaces using filters.

Types of Access Lists

You can create the following IPX access lists to filter various kinds of traffic:

•Standard access list—Restricts traffic based on the source network number. You can further restrict traffic by specifying a destination address and a source and destination address mask. Standard IPX access lists use numbers from 800 to 899 or names to identify them.

•Extended access list—Restricts traffic based on the IPX protocol type. You can further restrict traffic by specifying source and destination addresses and address masks, and source and destination sockets. Extended IPX access lists use numbers from 900 to 999 or names to identify them.

•SAP access list—Restricts traffic based on the IPX SAP type. These lists are used for SAP filters and GNS response filters. Novell SAP access lists use numbers from 1000 to 1099 or names to identify them.

•IPX NetBIOS access list—Restricts IPX NetBIOS traffic based on NetBIOS names, not numbers.

Types of Filters

There are more than 14 different IPX filters that you can define for IPX interfaces. They fall into the following six groups:

•Generic filters—Control which data packets are routed in or out of an interface based on the source and destination addresses and IPX protocol type of the packet.

•Routing table filters—Control which RIP updates are accepted and advertised by the Cisco IOS software, and from which devices the local router accepts RIP updates.

•SAP filters—Control which SAP services the Cisco IOS software accepts and advertises and which GNS response messages it sends out.

•IPX NetBIOS filters—Control incoming and outgoing IPX NetBIOS packets.

•Broadcast filters—Control which broadcast packets are forwarded.

Table 2 summarizes the filters, the access lists they use, and the commands used to define the filters in the first five groups. Use the show ipx interfaces command to display the filters defined on an interface.

ipx access-group 
{access-list-number | name} [in | out] 

ipx input-network-filter 
{access-list-number | name} 

ipx output-network-filter 
{access-list-number | name} 

distribute-list {access-list-number | name} 
out [interface-name | routing-process]

ipx router-filter 
{access-list-number | name} 

ipx input-sap-filter 
{access-list-number | name} 

ipx output-sap-filter 
{access-list-number | name} 

ipx router-sap-filter 
{access-list-number | name} 

ipx output-gns-filter 
{access-list-number | name} 

ipx netbios input-access-filter host name 

ipx netbios input-access-filter bytes name 

ipx netbios output-access-filter host name 

ipx netbios output-access-filter bytes name 

ipx helper-list {access-list-number | name}

Implementation Considerations

Remember the following information when configuring IPX network access control:

•Access lists entries are scanned in the order you enter them. The first matching entry is used. To improve performance, we recommend that you place the most commonly used entries near the beginning of the access list.

•An implicit deny everything entry is defined at the end of an access list unless you include an explicit permit everything entry at the end of the list.

•For numbered access lists, all new entries to an existing list are placed at the end of the list. You cannot add an entry to the middle of a list. Consequently, if you have previously included an explicit permit everything entry, new entries will never be scanned. The solution is to delete the access list and reenter it with the new entries.

For named access lists, all new entries to an existing list are placed at the end of the list. You cannot add entries to the middle of a list. However, you can remove specific entries using the no deny and no permit commands, rather than deleting the entire access list.

•Do not set up conditions that result in packets getting lost. One way you can lose packets is when a device or interface is configured to advertise services on a network that has access lists that deny these packets.

Controlling Access to IPX Networks Task List

To control access to IPX networks, perform the required tasks in the following sections:

•Creating Access Lists (Required)

•Creating Filters (Required)

Creating Access Lists

You can create access lists using numbers or names. You can choose which method you prefer. If you use numbers to identify your access lists, you are limited to 100 access lists per filter type. If you use names to identify your access lists, you can have an unlimited number of access lists per filter type.

The following sections describe how to perform these tasks:

•Creating Access Lists Using Numbers (Optional)

•Creating Access Lists Using Names (Optional)

Creating Access Lists Using Numbers

To create access lists using numbers, use one or more of the following commands in global configuration mode:

Once you have created an access list using numbers, apply it to the appropriate interfaces using filters as described in the "Creating Filters" section later in this chapter. Applying a filter will activate the access list.

Creating Access Lists Using Names

IPX named access lists allow you to identify IPX access lists with an alphanumeric string (a name) rather than a number. Using IPX named access lists allows you to maintain security by using a separate and easily identifiable access list for each user or interface. IPX named access lists also remove the limit of 100 lists per filter type.You can configure an unlimited number of the following types of IPX named access lists:

•Standard

•Extended

•SAP

•NetBIOS

If you identify your access list with a name rather than a number, the mode and command syntax are slightly different.

Implementation Considerations

Consider the following information before configuring IPX named access lists:

•Except for NetBIOS access lists, access lists specified by name are not compatible with releases prior to Cisco IOS Release 11.2(4)F.

•Access list names must be unique across all protocols.

•Except for NetBIOS access lists, numbered access lists are also available.

IPX Named Access List Configuration Task List

To configure IPX named access lists for standard, extended, SAP, NLSP route aggregation (summarization), or NetBIOS access lists, perform one or more of the tasks in the following sections:

•Creating a Named Standard Access List (Optional)

•Creating a Named Extended Access List (Optional)

•Creating a Named SAP Filtering Access List (Optional)

•Creating a NetBIOS Access List (Optional)

•Applying Time Ranges to Access Lists (Optional)

Creating a Named Standard Access List

To create a named standard access list, use the following commands beginning in global configuration mode:

For an example of creating a named standard access list, see the "Standard Named Access List Example" section at the end of this chapter.

Creating a Named Extended Access List

To create a named extended access list, use the following commands beginning in global configuration mode:

Creating a Named SAP Filtering Access List

To create a named access list for filtering SAP requests, use the following commands beginning in global configuration mode:

Creating a NetBIOS Access List

To create a NetBIOS access list, use one or more of the following commands in global configuration mode:

Modifying IPX Named Access Lists

After you initially create an access list, you place any subsequent additions (possibly entered from the terminal) at the end of the list. In other words, you cannot selectively add access list command lines to the middle of a specific access list. However, you can use no permit and no deny commands to remove entries from a named access list.

Note When creating access lists, remember that, by default, the end of the access list contains an implicit deny statement for everything if it did not find a match before reaching the end.

For an example of creating a generic filter, see the "IPX Network Access Examples" section at the end of this chapter.

Applying Named Access Lists to Interfaces

After creating an access list, you must apply it to the appropriate interface using filters as described in the "Creating Filters" section later in this chapter. Applying a filter will activate the access list.

Applying Time Ranges to Access Lists

It is now possible to implement access lists based on the time of day and week using the time-range command. To do so, first define the name of the time range and times of the day and week, then reference the time range by name in an access list to apply the restrictions of the time range to the access list.

Currently, IP and IPX named or numbered extended access lists are the only functions that can use time ranges. The time range allows the network administrator to define when the permit or deny statements in the access list are in effect. Prior to this time range feature, access list statements were always in effect once they were applied. The time-range keyword and argument are referenced in the named and numbered extended access list task tables in the previous sections, "Creating Access Lists Using Numbers" and "Creating Access Lists Using Names." The time-range command is configured in the "Performing Basic System Management" chapter of the Cisco IOS Configuration Fundamentals Configuration Guide. See the "IPX Network Access Examples"  section at the end of this chapter for a configuration example of IPX time ranges.

There are many possible benefits of time ranges, such as the following:

•The network administrator has more control over permitting or denying a user access to resources. These resources could be an application (identified by an IP address/mask pair and a port number), policy routing, or an on-demand link (identified as interesting traffic to the dialer).

•Network administrators can set time-based security policy, including:

–Perimeter security using the Cisco IOS Firewall feature set or access lists

–Data confidentiality with Cisco Encryption Technology or IPS

•Policy-based routing and queueing functions are enhanced.

•When provider access rates vary by time of day, it is possible to automatically reroute traffic cost effectively.

•Service providers can dynamically change a committed access rate (CAR) configuration to support the quality of service (QoS) Service Level Agreements (SLAs) that are negotiated for certain times of day.

•Network administrators can control logging messages. Access list entries can log traffic at certain times of the day, but not constantly. Therefore, administrators can simply deny access without needing to analyze many logs generated during peak hours.

Creating Filters

Filters allow you to control which traffic is forwarded or blocked at the interfaces of the router. Filters apply specific numbered or named access lists to interfaces.

To create filters, perform the tasks in the following sections:

•Creating Generic Filters (Optional)

•Creating Filters for Updating the Routing Table (Optional)

•Creating SAP Filters (Optional)

•Creating GNS Response Filters (Optional)

•Creating GGS Response Filters (Optional)

•Creating IPX NetBIOS Filters (Optional)

•Creating Broadcast Message Filters (Optional)

Creating Generic Filters

Generic filters determine which data packets to receive from or send to an interface, based on the source and destination addresses, IPX protocol type, and source and destination socket numbers of the packet.

To create generic filters, first create a standard or an extended access list as described in the "Creating Access Lists" section earlier in this chapter and then apply a filter to an interface.

To apply a generic filter to an interface, use the following command in interface configuration mode:

You can apply only one input filter and one output filter per interface or subinterface. You cannot configure an output filter on an interface where autonomous switching is already configured. Similarly, you cannot configure autonomous switching on an interface where an output filter is already present. You cannot configure an input filter on an interface if autonomous switching is already configured on any interface. Likewise, you cannot configure input filters if autonomous switching is already enabled on any interface.

For an example of creating a generic filter, see the "IPX Network Access Examples" section at the end of this chapter.

Creating Filters for Updating the Routing Table

Routing table update filters control the entries that the Cisco IOS software accepts for its routing table, and the networks that it advertises in its routing updates.

To create filters to control updating of the routing table, first create a standard or an extended access list as described in the "Creating Access Lists" section earlier in this chapter and then apply one or more routing filters to an interface.

To apply routing table update filters to an interface, use one or more of the following commands in interface configuration or router configuration mode:

Note The ipx output-network-filter command applies to the IPX RIP only. To control the advertising of routes when filtering routing updates in Enhanced IGRP, use the distribute-list out command. See the "Controlling the Advertising of Routes in Routing Updates" section earlier in this chapter for more information.

Creating SAP Filters

A common source of traffic on Novell networks is SAP messages, which are generated by NetWare servers and the Cisco IOS software when they broadcast their available services.

To control how SAP messages from network segments or specific servers are routed among IPX networks, first create a SAP filtering access list as described in the "Creating Access Lists" section earlier in this chapter and then apply one or more filters to an interface.

To apply SAP filters to an interface, use one or more of the following commands in interface configuration mode:

You can apply one of each SAP filter to each interface.

For examples of creating and applying SAP filters, see the "SAP Input Filter Example" and "SAP Output Filter Example" sections at the end of this chapter.

Creating GNS Response Filters

To create filters for controlling which servers are included in the GNS responses sent by the Cisco IOS software, first create a SAP filtering access list as described in the "Creating Access Lists" section earlier in this chapter and then apply a GNS filter to an interface.

To apply a GNS filter to an interface, use the following command in interface configuration mode:

Creating GGS Response Filters

To create filters for controlling which servers are included in the Get General Service (GGS) responses sent by the Cisco IOS software, first create a SAP filtering access list as described in the "Creating Access Lists" section earlier in this chapter and then apply a GGS filter to an interface.

Note Because GGS SAP response filters are applied ahead of output SAP filters, a SAP entry permitted to pass through the GGS SAP response filter can still be filtered by the output SAP filter.

To apply a GGS filter to an interface, use the following command in interface configuration mode:

For an example of creating a GGS SAP response filter, see the "IPX Network Access Examples" section at the end of this chapter.

Creating IPX NetBIOS Filters

The Novell IPX NetBIOS allows messages to be exchanged between nodes using alphanumeric names and node addresses. Therefore, the Cisco IOS software lets you filter incoming and outgoing NetBIOS FindName packets by the node name or by an arbitrary byte pattern (such as the node address) in the packet.

Note These filters apply to IPX NetBIOS FindName packets only. They have no effect on Logic Link Control, type 2 (LLC2) NetBIOS packets.

Implementation Considerations

Remember the following when configuring IPX NetBIOS access control:

•Host (node) names are case sensitive.

•Host and byte access lists can have the same names because the two types of lists are independent of each other.

•When nodes are filtered by name, the names in the access lists are compared with the destination name field for IPX NetBIOS "find name" requests.

•Access filters that filter by byte offset can have a significant impact on the packet transmission rate because each packet must be examined. You should use these access lists only when absolutely necessary.

•If a node name is not found in an access list, the default action is to deny access.

Configuring IPX NetBIOS Filters

To create filters for controlling IPX NetBIOS access, first create a NetBIOS access list as described in the "Creating Access Lists"  section earlier in this chapter and then apply the access list to an interface.

To apply a NetBIOS access list to an interface, use one or more of the following commands in interface configuration mode:

You can apply one of each of these four filters to each interface.

For an example of how to create filters for controlling IPX NetBIOS, see the "IPX NetBIOS Filter Examples" section at the end of this chapter.

Creating Broadcast Message Filters

Routers normally block all broadcast requests and do not forward them to other network segments, therefore preventing the degradation of performance inherent in broadcast traffic over the entire network. You can define which broadcast messages get forwarded to other networks by applying a broadcast message filter to an interface.

To create filters for controlling broadcast messages, first create a standard or an extended access list as described in the "Creating Access Lists" section earlier in this chapter and then apply a broadcast message filter to an interface.

To apply a broadcast message filter to an interface, use the following commands in interface configuration mode:

Note A broadcast message filter has no effect unless you have issued an ipx helper-address or an ipx type-20-propagation command on the interface to enable and control the forwarding of broadcast messages. These commands are discussed later in this chapter.

For examples of creating and applying broadcast message filters, see the "Helper Facilities to Control Broadcast Examples" section at the end of this chapter.

Tuning IPX Network Performance

To tune IPX network performance, perform the tasks in one or more of the following sections:

•Controlling Novell IPX Compliance (Optional)

•Adjusting RIP and SAP Information (Optional)

•Configuring Load Sharing (Optional)

•Specifying the Use of Broadcast Messages (Optional)

•Disabling IPX Fast Switching (Optional)

•Adjusting the Route Cache (Optional)

•Adjusting Default Routes (Optional)

•Padding Odd-Length Packets (Optional)

Controlling Novell IPX Compliance

The Cisco implementation of the Novell IPX protocol is certified to provide full IPX router functionality, as defined by the Novell IPX Router Specification, version 1.10 publication published November 17, 1992.

To control compliance to Novell specifications, perform the tasks in the following sections:

•Controlling the Forwarding of Type 20 Packets (Optional)

•Controlling Interpacket Delay (Optional)

•Shutting Down an IPX Network (Optional)

•Achieving Full Novell Compliance (Optional)

Controlling the Forwarding of Type 20 Packets

NetBIOS over IPX uses Type 20 propagation broadcast packets flooded to all networks to get information about the named nodes on the network. NetBIOS uses a broadcast mechanism to get this information, because it does not implement a network layer.

Routers normally block all broadcast requests. By enabling Type 20 packet propagation, IPX interfaces on the router may accept and forward Type 20 packets.

How Type 20 Packet Propagation Works

When an interface configured for Type 20 propagation receives a Type 20 packet, Cisco IOS software processes the packet according to Novell specifications. Cisco IOS software propagates the packet to the next interface. The Type 20 packet can be propagated for up to eight hop counts.

Loop Detection and Other Checks

Before forwarding (flooding) the packets, the router performs loop detection as described by the IPX router specification.

You can configure the Cisco IOS software to apply extra checks to Type 20 propagation packets above and beyond the loop detection described in the IPX specification. These checks are the same ones that are applied to helpered all-nets broadcast packets. They can limit unnecessary duplication of Type 20 broadcast packets. The extra helper checks are as follows:

•Accept Type 20 propagation packets only on the primary network, which is the network that is the primary path back to the source network.

•Forward Type 20 propagation packets only via networks that do not lead back to the source network.

Although this extra checking increases the robustness of Type 20 propagation packet handling by decreasing the amount of unnecessary packet replication, it has the following two side effects:

•If Type 20 packet propagation is not configured on all interfaces, these packets might be blocked when the primary interface changes.

•It might be impossible to configure an arbitrary, manual spanning tree for Type 20 packet propagation.

Relationship Between Type 20 Propagation and Helper Addresses

You use helper addresses to forward non-Type 20 broadcast packets to other network segments. For information on forwarding other broadcast packets, see the "Using Helper Addresses to Forward Broadcast Packets" section later in this chapter.

You can use helper addresses and Type 20 propagation together in your network. Use helper addresses to forward non-Type 20 broadcast packets and use Type 20 propagation to forward Type 20 broadcast packets.

Type 20 Packets Configuration Task List

You can enable the forwarding of Type 20 packets on individual interfaces. Additionally, you can restrict the acceptance and forwarding of Type 20 packets. You can also choose to not comply with Novell specifications and forward Type 20 packets using helper addresses rather than using Type 20 propagation. The following sections describe these tasks:

•Enabling the Forwarding of Type 20 Packets (Optional)

•Restricting the Acceptance of Incoming Type 20 Packets (Optional)

•Restricting the Forwarding of Outgoing Type 20 Packets (Optional)

•Forwarding Type 20 Packets Using Helper Addresses (Optional)

Enabling the Forwarding of Type 20 Packets

By default, Type 20 propagation packets are dropped by the Cisco IOS software. You can configure the software to receive Type 20 propagation broadcast packets and forward (flood) them to other network segments, subject to loop detection.

To enable the receipt and forwarding of Type 20 packets, use the following command in interface configuration mode:

When you enable Type 20 propagation, Cisco IOS propagates the broadcast to the next interface up to eight hops.

Restricting the Acceptance of Incoming Type 20 Packets

For incoming Type 20 propagation packets, the Cisco IOS software is configured by default to accept packets on all interfaces enabled to receive Type 20 propagation packets. You can configure the software to accept packets only from the single network that is the primary route back to the source network, which means that similar packets from the same source that are received via other networks will be dropped.

Checking of incoming Type 20 propagation broadcast packets is done only if the interface is configured to receive and forward Type 20 packets.

To impose restrictions on the receipt of incoming Type 20 propagation packets in addition to the checks defined in the IPX specification, use the following command in global configuration mode:

Restricting the Forwarding of Outgoing Type 20 Packets

For outgoing Type 20 propagation packets, the Cisco IOS software is configured by default to send packets on all interfaces enabled to send Type 20 propagation packets, subject to loop detection. You can configure the software to send these packets only to networks that are not routes back to the source network. (The software uses the current routing table to determine routes.)

Checking of outgoing Type 20 propagation broadcast packets is done only if the interface is configured to receive and forward Type 20 packets.

To impose restrictions on the transmission of Type 20 propagation packets, and to forward these packets to all networks using only the checks defined in the IPX specification, use the following command in global configuration mode:

Forwarding Type 20 Packets Using Helper Addresses

You can also forward Type 20 packets to specific network segments using helper addresses rather than using the Type 20 packet propagation.

You may want to forward Type 20 packets using helper addresses when some routers in your network are running versions of Cisco IOS that do not support Type 20 propagation. When some routers in your network support Type 20 propagation and others do not, you can avoid flooding packets everywhere in the network by using helper addresses to direct packets to certain segments only.

Cisco IOS Release 9.1 and earlier versions do not support Type 20 propagation.

Note Forwarding Type 20 packets using helper addresses does not comply with the Novell IPX router specification.

To forward Type 20 packets addresses using helper addresses, use the following commands beginning in global configuration mode:

The Cisco IOS software forwards Type 20 packets to only those nodes specified by the ipx helper-address command.

Note Using the ipx type-20-helpered command disables the receipt and forwarding of Type 20 propagation packets as directed by the ipx type-20-propagation command.

Controlling Interpacket Delay

To control interpacket delay, you can use a combination of global configuration and interface configuration commands.

Use one or more of the following commands in global configuration mode:

Use one or more of the following commands in interface configuration mode:

Note We recommend that you use the ipx output-rip-delay and ipx output-sap-delay commands on slower speed WAN interfaces. The default delay for Cisco IOS Release 11.1 and later versions is 55 milliseconds.

Shutting Down an IPX Network

To shut down an IPX network using a Novell-compliant method, use the following command in interface configuration mode:

Convergence is faster when you shut down an IPX network using the ipx down command than when using the shutdown command.

Achieving Full Novell Compliance

To achieve full compliance on each interface configured for IPX, use the following commands in interface configuration mode:

You can also globally set interpacket delays for multiple-packet RIP and SAP updates to achieve full compliance, eliminating the need to set delays on each interface. To set these interpacket delays, use the following commands in global configuration mode:

Note The default delay for Cisco IOS Release 11.1 and later versions is 55 milliseconds.

Adjusting RIP and SAP Information

To adjust RIP and SAP information, perform one or more of the optional tasks in the following sections:

•Configuring Static Routes (Optional)

•Adjusting the RIP Delay Field (Optional)

•Controlling Responses to RIP Requests (Optional)

•Adjusting RIP Update Timers (Optional)

•Configuring RIP Update Packet Size (Optional)

•Configuring Static SAP Table Entries (Optional)

•Configuring the Queue Length for SAP Requests (Optional)

•Adjusting SAP Update Timers (Optional)

•Configuring SAP Update Packet Size (Optional)

•Enabling SAP-after-RIP (Optional)

•Disabling Sending of General RIP or SAP Queries (Optional)

•Controlling Responses to GNS Requests (Optional)

Configuring Static Routes

IPX uses RIP, Enhanced IGRP, or NLSP to determine the best path when several paths to a destination exist. The routing protocol then dynamically updates the routing table. However, you might want to add static routes to the routing table to explicitly specify paths to certain destinations. Static routes always override any dynamically learned paths.

Be careful when assigning static routes. When links associated with static routes are lost, traffic may stop being forwarded or traffic may be forwarded to a nonexistent destination, even though an alternative path might be available.

To add a static route to the routing table, use the following command in global configuration mode:

You can configure static routes that can be overridden by dynamically learned routes. These routes are referred to as floating static routes. You can use a floating static route to create a path of last resort that is used only when no dynamic routing information is available.

Note By default, floating static routes are not redistributed into other dynamic protocols.

To add a floating static route to the routing table, use the following command in global configuration mode:

Adjusting the RIP Delay Field

By default, all LAN interfaces have a RIP delay of 1 and all WAN interfaces have a RIP delay of 6. Leaving the delay at its default value is sufficient for most interfaces. However, you can adjust the RIP delay field by setting the tick count. To set the tick count, use the following command in interface configuration mode:

Controlling Responses to RIP Requests

To control responses to RIP requests, use the following command in interface configuration mode:

Adjusting RIP Update Timers

You can set the interval between IPX RIP updates on a per-interface basis. You can also specify the delay between the packets of a multiple-packet RIP update on a per-interface or global basis. Additionally, you can specify the delay between packets of a multiple-packet triggered RIP update on a per-interface or global basis.

You can set RIP update timers only in a configuration in which all routers are Cisco routers, or in which the IPX routers allow configurable timers. The timers should be the same for all devices connected to the same cable segment. The update value you choose affects internal IPX timers as follows:

•IPX routes are marked invalid if no routing updates are heard within three times the value of the update interval (3 * interval) and are advertised with a metric of infinity.

•IPX routes are removed from the routing table if no routing updates are heard within four times the value of the update interval (4 * interval).

•If you define a timer for more than one interface in a router, the granularity of the timer is determined by the lowest value defined for one of the interfaces in the router. The router "wakes up" at this granularity interval and sends out updates as appropriate. For more information about granularity, refer to the "Novell IPX Commands" chapter in the Cisco IOS AppleTalk and Novell IPX Command Reference.

You might want to set a delay between the packets in a multiple-packet update if there are some slower PCs on the network or on slower-speed interfaces.

To adjust RIP update timers on a per-interface basis, use one or all of the following commands in interface configuration mode:

To adjust RIP update timers on a global basis, use one or both of the following commands in global configuration mode:

By default, the RIP entry for a network or server ages out at an interval equal to three times the RIP timer. To configure the multiplier that controls the interval, use the following command in interface configuration mode:

Configuring RIP Update Packet Size

By default, the maximum size of RIP updates sent out an interface is 432 bytes. This size allows for 50 routes at 8 bytes each, plus a 32-byte IPX RIP header. To modify the maximum packet size, use the following command in interface configuration mode:

Configuring Static SAP Table Entries

Servers use SAP to advertise their services via broadcast packets. The Cisco IOS software stores this information in the SAP table, also known as the Server Information Table. This table is updated dynamically. You might want to explicitly add an entry to the Server Information Table so that clients always use the services of a particular server. Static SAP assignments always override any identical entries in the SAP table that are learned dynamically, regardless of hop count. If a dynamic route that is associated with a static SAP entry is lost or deleted, the software will not announce the static SAP entry until it relearns the route.

To add a static entry to the SAP table, use the following command in global configuration mode:

Configuring the Queue Length for SAP Requests

The Cisco IOS software maintains a list of SAP requests to process, including all pending Get Nearest Server (GNS) queries from clients attempting to reach servers. When the network is restarted following a power failure or other unexpected event, the router can be inundated with hundreds of requests for servers. Typically, many of these are repeated requests from the same clients. You can configure the maximum length allowed for the pending SAP requests queue. SAP requests received when the queue is full are dropped, and the client must resend them.

To set the queue length for SAP requests, use the following command in global configuration mode:

Adjusting SAP Update Timers

You can adjust the interval at which SAP updates are sent. You can also set the delay between packets of a multiple-packet SAP update on a per-interface or global basis. Additionally, you can specify the delay between packets of a multiple-packet triggered SAP update on a per-interface or global basis.

Changing the interval at which SAP updates are sent is most useful on limited-bandwidth, point-to-point links such as slower-speed interfaces. You should ensure that all IPX servers and routers on a given network have the same SAP interval. Otherwise, they might decide that a server is down when it is really up.

It is not possible to change the interval at which SAP updates are sent on most PC-based servers. Therefore, you should never change the interval for an Ethernet or Token Ring network that has servers on it.

You can set the router to send an update only when changes have occurred. Using the changes-only keyword specifies the sending of a SAP update only when the link comes up, when the link is downed administratively, or when the databases change. The changes-only keyword causes the router to do the following:

•Send a single, full broadcast update when the link comes up

•Send appropriate triggered updates when the link is shut down

•Send appropriate triggered updates when specific service information changes

To modify the SAP update timers on a per-interface basis, use one or all of the following commands in interface configuration mode:

To adjust SAP update timers on a global basis (eliminating the need to configure delays on a per-interface basis), use one or both of the following commands in global configuration mode:

By default, the SAP entry of a network or server ages out at an interval equal to three times the SAP update interval. To configure the multiplier that controls the interval, use the following command in interface configuration mode:

Configuring SAP Update Packet Size

By default, the maximum size of SAP updates sent out on an interface is 480 bytes. This size allows for seven servers (64 bytes each), plus a 32-byte IPX SAP header. To modify the maximum packet size, use the following command in interface configuration mode:

Enabling SAP-after-RIP

The IPX SAP-after-RIP feature links SAP updates to RIP updates so that SAP broadcast and unicast updates automatically occur immediately after the completion of the corresponding RIP update. This feature ensures that a remote router does not reject service information because it lacks a valid route to the service. As a result of this feature, periodic SAP updates are sent at the same interval as RIP updates.

The default behavior of the router is to send RIP and SAP periodic updates with each using its own update interval, depending on the configuration. In addition, RIP and SAP periodic updates are jittered slightly, such that they tend to diverge from each other over time. This feature synchronizes SAP and RIP updates.

Sending all SAP and RIP information in a single update reduces bandwidth demands and eliminates erroneous rejections of SAP broadcasts.

Linking SAP and RIP updates populates the service table of the remote router more quickly, because services will not be rejected due to the lack of a route to the service. Populating the service table more quickly can be especially useful on WAN circuits where the update intervals have been greatly increased to reduce the overall level of periodic update traffic on the link.

To configure the router to send a SAP update following a RIP broadcast, use the following command in interface configuration mode:

Disabling Sending of General RIP or SAP Queries

You can disable the sending of general RIP or SAP queries on a link when it first comes up to reduce traffic and save bandwidth.

RIP and SAP general queries are normally sent by remote routers when a circuit first comes up. On WAN circuits, two full updates of each kind are often sent across the link. The first update is a full broadcast update, triggered locally by the link-up event. The second update is a specific (unicast) reply triggered by the general query received from the remote router. If you disable the sending of general queries when the link first comes up, it is possible to reduce traffic to a single update, and save bandwidth.

To disable the sending of a general RIP or SAP query when an interface comes up, use the following command in interface configuration mode:

To reenable the sending of a general RIP or SAP query, use the positive form of the command.

Controlling Responses to GNS Requests

You can set the method in which the router responds to SAP GNS requests, you can set the delay time in responding to these requests, or you can disable the sending of responses to these requests altogether.

By default, the router responds to GNS requests if appropriate. For example, if a local server with a better metric exists, then the router does not respond to the GNS request on that segment.

The default method of responding to GNS requests is to respond with the server whose availability was learned most recently.

To control responses to GNS requests, use one or both of the following commands in global configuration mode:

Note The ipx gns-response-delay command is also supported as an interface configuration command. To override the global delay value for a specific interface, use the ipx gns-response-delay command in interface configuration mode.

To disable GNS queries on a per-interface basis, use the following command in interface configuration mode:

Configuring Load Sharing

To configure IPX to perform round-robin or per-host load sharing, perform the tasks described in the following sections:

•Enabling Round-Robin Load Sharing (Optional)

•Enabling per-Host Load Sharing (Optional)

Enabling Round-Robin Load Sharing

You can set the maximum number of equal-cost, parallel paths to a destination. (Note that when paths have differing costs, the Cisco IOS software chooses lower-cost routes in preference to higher-cost routes.) The software then distributes output on a packet-by-packet basis in round-robin fashion. That is, the first packet is sent along the first path, the second packet along the second path, and so on. When the final path is reached, the next packet is sent to the first path, the next to the second path, and so on. This round-robin scheme is used regardless of whether fast switching is enabled.

Limiting the number of equal-cost paths can save memory on routers with limited memory or very large configurations. Additionally, in networks with a large number of multiple paths and systems with limited ability to cache out-of-sequence packets, performance might suffer when traffic is split between many paths.

To set the maximum number of paths, use the following command in global configuration mode:

Enabling per-Host Load Sharing

Round-robin load sharing is the default behavior when you configure ipx maximum-paths to a value greater than 1. Round-robin load sharing works by sending data packets over successive equal cost paths without regard to individual end hosts or user sessions. Path utilization increases transmission speed, but, because packets destined for a given end host may take different paths, they might arrive out of order.

You can address the possibility of packets arriving out of order by enabling per-host load sharing. With per-host load sharing, the router still uses multiple, equal-cost paths to achieve load sharing; however, packets for a given end host are guaranteed to take the same path, even if multiple, equal-cost paths are available. Traffic for different end hosts tend to take different paths, but true load balancing is not guaranteed. The exact degree of load balancing achieved depends on the exact nature of the workload.

To enable per-host load sharing, use the following commands in global configuration mode:

Specifying the Use of Broadcast Messages

To specify the use of broadcast messages, perform the tasks described in the following sections:

•Using Helper Addresses to Forward Broadcast Packets (Optional)

•Enabling Fast Switching of IPX Directed Broadcast Packets (Optional)

Using Helper Addresses to Forward Broadcast Packets

Routers normally block all broadcast requests and do not forward them to other network segments, therefore preventing the degradation of performance over the entire network. However, you can enable the router to forward broadcast packets to helper addresses on other network segments.

How Helper Addresses Work

Helper addresses specify the network and node on another segment that can receive unrecognized broadcast packets. Unrecognized broadcast packets are non-RIP and non-SAP packets that are not addressed to the local network.

When the interface configured with helper addresses receives an unrecognized broadcast packet, Cisco IOS software changes the broadcast packet to a unicast and sends the packet to the specified network and node on the other network segment. Unrecognized broadcast packets are not flooded everywhere in your network.

With helper addresses, there is no limit on the number of hops that the broadcast packet can make.

Fast Switching Support

Cisco IOS supports fast switching of helpered broadcast packets.

When to Use Helper Addresses

You use helper addresses when you want to forward broadcast packets (except Type 20 packets) to other network segments.

Forwarding broadcast packets to helper addresses is sometimes useful when a network segment does not have an end-host capable of servicing a particular type of broadcast request. You can specify the address of a server, network, or networks that can process the broadcast packet.

Relationship Between Helper Addresses and Type 20 Propagation

You use Type 20 packet propagation to forward Type 20 packets to other network segments. For information on forwarding Type 20 packets, see the "Controlling the Forwarding of Type 20 Packets" section earlier in this chapter.

You can use helper addresses and Type 20 propagation together in your network. Use helper addresses to forward non-Type 20 broadcast packets and use Type 20 propagation to forward Type 20 broadcast packets.

Implementation Considerations

Using helper addresses is not Novell-compliant. However, it does allow routers to forward broadcast packets to network segments that can process them without flooding the network. It also allows routers running versions of Cisco IOS that do not support Type 20 propagation to forward Type 20 packets.

The Cisco IOS software supports all-networks flooded broadcasts (sometimes referred to as all-nets flooding). These are broadcast messages that are forwarded to all networks. Use all-nets flooding carefully and only when necessary, because the receiving networks may be overwhelmed to the point that no other traffic can traverse them.

Use the ipx helper-list command, described earlier in this chapter, to define access lists that control which broadcast packets get forwarded.

Using Helper Addresses

To specify a helper address for forwarding broadcast packets, use the following command in interface configuration mode:

You can specify multiple helper addresses on an interface.

For an example of using helper addresses to forward broadcast messages, see the "Helper Facilities to Control Broadcast Examples" section at the end of this chapter.

Enabling Fast Switching of IPX Directed Broadcast Packets

By default, Cisco IOS software switches packets that have been helpered to the broadcast address. To enable fast switching of these IPX-directed broadcast packets, use the following command in global configuration mode:

Disabling IPX Fast Switching

By default, fast switching is enabled on all interfaces that support fast switching.

Fast switching allows higher throughput by switching a packet using a cache created by previous packets. Fast switching is enabled by default on all interfaces that support fast switching.

Packet transfer performance is generally better when fast switching is enabled. However, you might want to disable fast switching in order to save memory space on interface cards and to help avoid congestion when high-bandwidth interfaces are writing large amounts of information to low-bandwidth interfaces.

Caution Turning off fast switching increases system overhead.

To disable IPX fast switching, use the following command in interface configuration mode:

Adjusting the Route Cache

Adjusting the route cache allows you to control the size of the route cache, reduce memory consumption, and improve router performance. You accomplish these tasks by controlling the route cache size and invalidation. The following sections describe these optional tasks:

•Controlling Route Cache Size (Optional)

•Controlling Route Cache Invalidation (Optional)

Controlling Route Cache Size

You can limit the number of entries stored in the IPX route cache to free up router memory and aid router processing.

Storing too many entries in the route cache can use a significant amount of router memory, causing router processing to slow. This situation is most common on large networks that run network management applications for NetWare.

For example, if a network management station is responsible for managing all clients and servers in a very large (greater than 50,000 nodes) Novell network, the routers on the local segment can become inundated with route cache entries. You can set a maximum number of route cache entries on these routers to free up router memory and aid router processing.

To set a maximum limit on the number of entries in the IPX route cache, use the following command in global configuration mode:

If the route cache has more entries than the specified limit, the extra entries are not deleted. However, they may be removed if route cache invalidation is in use. See the "Controlling Route Cache Invalidation" section later in this chapter for more information on invalidating route cache entries.

Controlling Route Cache Invalidation

You can configure the router to invalidate fast-switch cache entries that are inactive. If these entries remain invalidated for 1 minute, the router purges the entries from the route cache.

Purging invalidated entries reduces the size of the route cache, reduces memory consumption, and improves router performance. Also, purging entries helps ensure accurate route cache information.

You specify the period of time that valid fast-switch cache entries must be inactive before the router invalidates them. You can also specify the number of cache entries that the router can invalidate per minute.

To configure the router to invalidate fast-switch cache entries that are inactive, use the following command in global configuration mode:

When you use the ipx route-cache inactivity-timeout command with the ipx route-cache max-size command, you can ensure a small route cache with fresh entries.

Adjusting Default Routes

You can adjust the use of default routes in your IPX network. You can turn off the use of network number -2 as the default route. You can also specify that the router advertise only default RIP routes out an interface. The following sections describe these optional tasks:

•Disabling Network Number -2 as the Default Route (Optional)

•Advertising Only Default RIP Routes (Optional)

Disabling Network Number -2 as the Default Route

The default route is used when a route to any destination network is unknown. All packets for which a route to the destination address is unknown are forwarded to the default route. By default, IPX treats network number -2 (0xFFFFFFFE) as the default route.

For an introduction to default routes, see the "IPX Default Routes" section earlier in this chapter. For more background information on how to handle IPX default routes, refer to the Novell NetWare Link Services Protocol (NLSP) Specification, Revision 1.1 publication.

By default, Cisco IOS software treats network -2 as the default route. You can disable this default behavior and use network -2 as a regular network number in your network.

To disable the use of network number -2 as the default route, use the following command in global configuration mode:

Advertising Only Default RIP Routes

Unless configured otherwise, all known RIP routes are advertised out each interface. However, you can choose to advertise only the default RIP route if it is known, therefore greatly reducing the CPU overhead when routing tables are large.

To advertise only the default route via an interface, use the following command in interface configuration mode:

Padding Odd-Length Packets

Some IPX end hosts accept only even-length Ethernet packets. If the length of a packet is odd, the packet must be padded with an extra byte so that end host can receive it. By default, Cisco IOS pads odd-length Ethernet packets.

However, there are cases in certain topologies where nonpadded Ethernet packets are forwarded onto a remote Ethernet network. Under specific conditions, you can enable padding on intermediate media as a temporary workaround for this problem. Note that you should perform this task only under the guidance of a customer engineer or other service representative.

To enable the padding of odd-length packets, use the following commands in interface configuration mode:

Shutting Down an IPX Network

You can administratively shut down an IPX network in two ways. In the first way, the network still exists in the configuration, but is not active. When shutting down, the network sends out update packets informing its neighbors that it is shutting down, therefore allowing the neighboring systems to update their routing, SAP, and other tables without needing to wait for routes and services learned via this network to time out.

To shut down an IPX network such that the network still exists in the configuration, use the following command in interface configuration mode:

To shut down an IPX network and remove it from the configuration, use one of the following commands in interface configuration mode:

When multiple networks are configured on an interface and you want to shut down one of the secondary networks and remove it from the interface, use the second command in the previous table specifying the network number of one of the secondary networks.

For an example of shutting down an IPX network, see the "IPX Routing Examples" section at the end of this chapter.

Configuring IPX Accounting

IPX accounting enables you to collect information about IPX packets and the number of bytes that are switched through the Cisco IOS software. You collect information based on the source and destination IPX address. IPX accounting tracks only IPX traffic that is routed out an interface on which IPX accounting is configured; it does not track traffic generated by or terminated at the router itself.

The Cisco IOS software maintains two accounting databases: an active database and a checkpoint database. The active database contains accounting data tracked until the database is cleared. When the active database is cleared, its contents are copied to the checkpoint database. Using these two databases together enables you to monitor both current traffic and traffic that has previously traversed the router.

Switching Support

Process and fast switching support IPX accounting statistics. Autonomous and silicon switching engine (SSE) switching do not support IPX accounting statistics.

Note CiscoBus (Cbus) and SSE are not supported on the MIP interface.

Access List Support

IPX access lists support IPX accounting statistics.

IPX Accounting Task List

To configure IPX accounting, perform the tasks in the following sections. The first task is required; the remaining task is optional.

•Enabling IPX Accounting (Required)

•Customizing IPX Accounting (Optional)

Enabling IPX Accounting

To enable IPX accounting, use the following command in global configuration mode:

Customizing IPX Accounting

To customize IPX accounting, use one or more of the following commands in global configuration mode:

Transit entries are entries in the database that do not match any of the networks specified by the ipx accounting-list commands.

If you enable IPX accounting on an interface but do not specify an accounting list, IPX accounting tracks all traffic through the interface (all transit entries) up to the accounting threshold limit.

For an example of how to configure IPX accounting, see the "IPX Accounting Example" section at the end of this chapter.

Configuring IPX Between LANs

Cisco IOS software supports routing IPX between Ethernet-emulated LANs and Token Ring-emulated LANs. For more information on emulated LANs and routing IPX between them, refer to the "Configuring LAN Emulation" chapter of the Cisco IOS Switching Services Configuration Guide.

Configuring IPX Between VLANs

Cisco IOS software supports routing IPX between VLANs. Users with Novell NetWare environments can configure any one of the four IPX Ethernet encapsulations to be routed using the Inter-Switch Link (ISL) encapsulation across VLAN boundaries. For more information on VLANs and routing IPX between them over ISL, refer to the "Configuring Routing Between VLANs with ISL Encapsulation" chapter of the Cisco IOS Switching Services Configuration Guide.

Configuring IPX Multilayer Switching

Cisco IOS software supports IPX Multilayer Switching (MLS). For more information on IPX MLS, refer to the "Multilayer Switching" chapter of the Cisco IOS Switching Services Configuration Guide.

Monitoring and Maintaining the IPX Network

To monitor and maintain your IPX network, perform the optional tasks described in the following sections:

•General Monitoring and Maintaining Tasks (Optional)

•Monitoring and Maintaining IPX Enhanced IGRP (Optional)

•Monitoring and Maintaining IPX Accounting (Optional)

General Monitoring and Maintaining Tasks

You can perform one or more of these general monitoring and maintaining tasks as described in the following sections:

•Monitoring and Maintaining Caches, Tables, Interfaces, and Statistics (Optional)

•Specifying the Type and Use of Ping Packets (Optional)

•Troubleshooting Network Connectivity (Optional)

Monitoring and Maintaining Caches, Tables, Interfaces, and Statistics

To monitor and maintain caches, tables, interfaces, or statistics in a Novell IPX network, use one or more of the following commands in EXEC mode:

Specifying the Type and Use of Ping Packets

The Cisco IOS software can send Cisco pings and standard Novell pings as defined in the NLSP specification or diagnostic request packets. By default, the software generates Cisco pings. To choose the ping type, use the following command in global configuration mode:

The IPX diagnostic ping feature addresses diagnostic related issues by accepting and processing unicast or broadcast diagnostic packets. It makes enhancements to the current IPX ping command to ping other stations using the diagnostic packets and display the configuration information in the response packet.

Note When a ping is sent from one station to another, the response is expected to come back immediately; when the ipx ping-default command is set to diagnostics, the response could consist of more than one packet and each node is expected to respond within 0.5 seconds of receipt of the request. Due to the absence of an end-of-message flag, there is a delay and the requester must wait for all responses to arrive. Therefore, in verbose mode there may be a brief delay of 0.5 seconds before the response data is displayed.

The ipx ping command using the diagnostic keyword can be used to conduct a reachability test and should not be used to measure accurate round-trip delay.

To initiate a ping, use one of the following commands in EXEC mode:

Troubleshooting Network Connectivity

To trace the IPX destination and measure roundtrip delays, use the following command in either user or privileged EXEC mode:

Note In user EXEC mode, you are not allowed to change the trace route timeout interval, probe count, minimum and maximum time to live, and verbose mode. To do so, use the trace command in privileged EXEC mode.

Monitoring and Maintaining IPX Enhanced IGRP

To monitor and maintain Enhanced IGRP on an IPX network, use one or more of the following commands in EXEC mode:

Logging Enhanced IGRP Neighbor Adjacency Changes

You can enable the logging of neighbor adjacency changes to monitor the stability of the routing system and to help you detect problems. By default, adjacency changes are not logged.

To enable logging of Enhanced IGRP neighbor adjacency changes, use the following command in IPX-router configuration mode:

Monitoring and Maintaining IPX Accounting

To monitor and maintain IPX accounting in your IPX network, use the following commands in EXEC mode:

Any Internet Protocol (IP) addresses used in this document are not intended to be actual addresses. Any examples, command display output, and figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses in illustrative content is unintentional and coincidental.

© 2007 Cisco Systems, Inc. All rights reserved.