|
|
Novell Internet Packet Exchange (IPX) is derived from the Xerox Network Systems (XNS) Internet Datagram Protocol (IDP). One major difference between IPX and XNS is that they do not always use the same Ethernet encapsulation format. A second difference is that IPX uses Novell's proprietary Service Advertisement Protocol (SAP) to advertise special network services.
Our implementation of Novell's IPX protocol has been certified as providing full IPX router functionality. Our router connects Ethernet, Token Ring, and FDDI networks, either directly or through high-speed serial lines (56 kbps to T1 speeds), X.25, or Frame Relay. The Cisco X.25 and T1 support currently is not compatible with Novell. This means that our routers must be used on both ends of T1 and X.25 circuits.
Use the commands in this chapter to configure and monitor Novell IPX networks. For IPX configuration information and examples, refer to the "Configuring Novell IPX" chapter in the Router Products Configuration Guide.
To define a standard IPX access list, use the standard version of the access-list global configuration command. To remove a standard access list, use the no form of this command.
access-list access-list-number {deny | permit} source-network[.source-node| access-list-number | Number of the access list. This is a decimal number from 800 to 899. |
| deny | Denies access if the conditions are matched. |
| permit | Permits access if the conditions are matched. |
| source-network | Number of the network from which the packet is being sent. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. A network number of 0 matches the local network. A network number of -1 matches all networks.
You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
| source-node | (Optional) Node on source-network from which the packet is being sent. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
| source-node-mask | (Optional) Mask to be applied to source-node. This is a 48-bit value represented as a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). Place ones in the bit positions you want to mask. |
| destination-network | (Optional) Number of the network to which the packet is being sent. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. A network number of 0 matches the local network. A network number of -1 matches all networks.
You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
| destination-node | (Optional) Node on destination-network to which the packet is being sent. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
| destination-node-mask | (Optional) Mask to be applied to destination-node. This is a 48-bit value represented as a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). Place ones in the bit positions you want to mask. |
No access lists are predefined.
Global configuration
Standard IPX access lists filter on the source network. All other parameters are optional.
Use the ipx access-group command to assign an access list to an interface. You can apply only one extended or one standard access list to an interface. The access list filters all outgoing packets on the interface.
To delete a standard access list, specify the minimum number of keywords and arguments needed to delete the proper access list. For example, to delete the entire access list, use the following command:
no access-list access-list-numberTo delete the access list for a specific network, use the following command:
no access-list access-list-number {deny | permit} source-networkThe following example denies access to traffic from all IPX networks (-1) to destination network 2:
access-list 800 deny -1 2
The following example denies access to all traffic from IPX address 1.0000.0c00.1111:
access-list 800 deny 1.0000.0c00.1111
The following example denies access from all nodes on network 1 that have a source address beginning with 0000.0c:
access-list 800 deny 1.0000.0c00.0000 0000.00ff.ffff
The following example denies access from source address 1111.1111.1111 on network 1 to destination address 2222.2222.2222 on network 2:
access-list 800 deny 1.1111.1111.1111 0000.0000.0000 2.2222.2222.2222 0000.0000.0000
access-list 800 deny 1.1111.1111.1111 2.2222.2222.2222
A dagger (+) indicates that the command is documented in another chapter.
access-list (extended)
ipx access-group
ipx input-network-filter
ipx output-network-filter
ipx router-filter
priority-list protocol +
To define an extended Novell IPX access list, use the extended version of the access-list global configuration command. To remove an extended access list, use the no form of this command.
access-list access-list-number {deny | permit} protocol [source-network][[[.source-node]| access-list-number | Number of the access list. This is a decimal number from 900 to 999. |
| deny | Denies access if the conditions are matched. |
| permit | Permits access if the conditions are matched. |
| protocol | Number of an IPX protocol type, in decimal. This also is sometimes referred to as the packet type. Table 19-1 in the "Usage Guidelines" section lists some IPX protocol numbers. |
| source-network | (Optional) Number of the network from which the packet is being sent. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. A network number of 0 matches the local network. A network number of -1 matches all networks.
You do not need to specify leading zeros in the network number, e.g., for the network number 000000AA, enter AA. |
| source-node | (Optional) Node on source-network from which the packet is being sent. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
| source-network-mask | (Optional) Mask to be applied to source-network. This is an eight-digit hexadecimal mask. Place ones in the bit positions you want to mask.
The mask must immediately be followed by a period, which must in turn immediately be followed by source-node-mask. |
| source-node-mask | (Optional) Mask to be applied to source-node. This is a 48-bit value represented as a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). Place ones in the bit positions you want to mask. |
| source-socket | Socket number from which the packet is being sent, in hexadecimal. Table 19-2 in the "Usage Guidelines" section lists some IPX socket numbers. |
| destination-network | (Optional) Number of the network to which the packet is being sent. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. A network number of 0 matches the local network. A network number of -1 matches all networks.
You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
| destination-node | (Optional) Node on destination-network to which the packet is being sent. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
| destination-network-mask | (Optional) Mask to be applied to destination-network. This is an eight-digit hexadecimal mask. Place ones in the bit positions you want to mask.
The mask must immediately be followed by a period, which must in turn immediately be followed by destination-node-mask. |
| destination-node-mask | (Optional) Mask to be applied to destination-node. This is a
48-bit value represented as a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). Place ones in the bit positions you want to mask. |
| destination-socket | (Optional) Socket number to which the packet is being sent, in hexadecimal. Table 19-2 in the "Usage Guidelines" section lists some IPX socket numbers. |
No access lists are predefined.
Global configuration
Extended IPX access lists filter on protocol type. All other parameters are optional.
If a network mask is used, all other fields are required.
Use the ipx access-group command to assign an access list to an interface. You can apply only one extended or one standard access list to an interface. The access list filters all outgoing packets on the interface.
Table 19-1 lists some IPX protocol numbers. Table 19-2 lists some IPX socket numbers. For additional information about IPX protocol numbers and socket numbers, contact Novell.
| IPX Protocol Number (Decimal) | Protocol (Packet Type) |
|---|---|
| 0 | Any protocol; refer to the socket number to determine the packet type |
| 1 | Routing Information Protocol (RIP) |
| 4 | Service Advertisement Protocol (SAP) |
| 5 | Sequenced Packet Exchange (SPX) |
| 17 | NetWare Core Protocol (NCP) |
| 20 | IPX NetBIOS |
| IPX Socket Number (Hexadecimal) | Socket |
|---|---|
| 0 | All sockets, wild card used to match all sockets |
| 451 | NetWare Core Protocol (NCP) process |
| 452 | Service Advertisement Protocol (SAP) process |
| 453 | Routing Information Protocol (RIP) process |
| 455 | Novell NetBIOS process |
| 456 | Novell diagnostic packet |
| 457 | Novell serialization socket |
| 4000-7FFF | Dynamic sockets; used by workstations for interaction with file servers and other network servers |
| 8000-FFFF | Well-known sockets as assigned by Novell |
To delete an extended access list, specify the minimum number of keywords and arguments needed to delete the proper access list. For example, to delete the entire access list, use the following command:
no access-list access-list-numberTo delete the access list for a specific protocol, use the following command:
no access-list access-list-number {deny | permit} protocolThe following example denies access to all RIP packets (protocol number 1) from socket 453 (RIP process socket) on source network 1 that are destined for socket 453 on network 2. It permits all other traffic.
access-list 900 deny 1 453 2 453 access-list 900 permit 0 -1 0 -1 0
The following example permits type 2 packets from any socket on network 10 to access any sockets on any nodes on networks 1000 through 100F. It denies all other traffic (with an implicit deny all):
access-list 910 permit 20 10.0000.0C00.0000 0000.0000.FFFF 0 1000.0000.0000.0000 F.FFFF.FFFF.FFFF 0
A dagger (+) indicates that the command is documented in another chapter.
access-list (standard)
ipx access-group
ipx input-network-filter
ipx output-network-filter
ipx router-filter
priority-list protocol +
To define an access list for filtering Service Advertisement Protocol (SAP) requests, use the SAP filtering form of the access-list global configuration command. To remove the access list, use the no form of this command.
access-list access-list-number {deny | permit} network[.node] [network.node-mask]| access-list-number | Number of the SAP access list. This is a decimal number from 1000 to 1099. |
| deny | Denies access if the conditions are matched. |
| permit | Permits access if the conditions are matched. |
| network | Network number. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. A network number of 0 matches the local network. A network number of -1 matches all networks.
You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
| node | (Optional) Node on network. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
| network.node-mask | (Optional) Mask to be applied to network and node. Place ones in the bit positions to be masked. |
| service-type | (Optional) Service type on which to filter. This is a hexadecimal number. A value of 0 means all services. Table 19-3 in the "Usage Guidelines" section lists examples of service types. |
| server-name | (Optional) Name of the server providing the specified service type. This can be any contiguous string of printable ASCII characters. Use double quotation marks (" ") to enclose strings containing embedded spaces. You can use an asterisk (*) at the end of the name as a wildcard to match one or more trailing characters. |
No access lists are predefined.
Global configuration
When configuring SAP filters for NetWare 3.11 and later servers, use the server's internal network and node number (the node number is always 0000.0000.0001) as its address in the access-list command. Do not use the network.node address of the particular interface board.
Table 19-3 lists some sample IPX SAP types. For more information about SAP types, contact Novell.
| Service Type (Hexadecimal) | Description |
|---|---|
| 0 | All SAP services; IPX defines server type 0 to be an unknown service, which means that you cannot define an access list to permit or deny unknown services |
| 1 | User |
| 2 | User group |
| 3 | Print server queue |
| 4 | File server |
| 5 | Job server |
| 7 | Print server |
| 9 | Archive server |
| A | Queue for job servers |
| 21 | NAS SNA gateway |
| 2D | Time Synchronization VAP |
| 2E | Dynamic SAP |
| 47 | Advertising print server |
| 4B | Btrieve VAP 5.0 |
| 4C | SQL VAP |
| 7A | TES--NetWare for VMS |
| 98 | NetWare access server |
| 9A | Named Pipes server |
| 9E | Portable NetWare--UNIX |
| 111 | Test server |
| 166 | NetWare management |
| 26A | NetWare management |
| FFFF | Wildcard (any SAP service) |
To delete a SAP access list, specify the minimum number of keywords and arguments needed to delete the proper access list. For example, to delete the entire access list, use the following command:
no access-list access-list-numberTo delete the access list for a specific network, use the following command:
no access-list access-list-number {deny | permit} networkThe following access list blocks all access to a file server (service type 4) on the directly attached network by resources on other Novell networks, but allows access to all other available services on the interface:
access-list 1001 deny -1 4 access-list 1001 permit -1
A dagger (+) indicates that the command is documented in another chapter.
ipx input-sap-filter
ipx output-gns-filter
ipx output-sap-filter
ipx router-sap-filter
priority-list protocol +
To delete all entries in the accounting database when IPX accounting is enabled, use the clear ipx accounting EXEC command.
clear ipx accounting [checkpoint]| no argument | Deletes all entries in the active database. |
| checkpoint | (Optional) Clears the checkpointed database. |
EXEC
You can also delete all entries in the checkpointed database by issuing the clear ipx accounting command twice in succession.
The following example clears all entries in the active database:
clear ipx accounting
ipx accounting
ipx accounting-list
ipx accounting-threshold
ipx accounting-transits
show ipx accounting
To delete entries from the IPX fast-switching cache, use the clear ipx cache EXEC command.
clear ipx cacheThis command has no arguments or keywords.
EXEC
The clear ipx cache command clears entries used for fast switching, autonomous switching, and SSE fast switching.
The following example deletes all entries from the IPX fast-switching cache:
clear ipx cache
ipx route-cache
show ipx cache
To delete routes from the IPX routing table, use the clear ipx route EXEC command.
clear ipx route [network | *]| network | (Optional) Number of the network whose routing table entry you want to delete. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
| * | (Optional) Deletes all routes in the routing table. |
EXEC
The following example clears the entry for network 3 from the IPX routing table:
clear ipx route 3
show ipx route
To have the Cisco 7000 series route processor (RP) recompute the entries in the IPX SSE fast-switching cache, use the clear ipx sse EXEC command.
clear ipx sseThis command has no arguments or keywords.
EXEC
Recomputing the entries in the RP's SSE fast-switching cache also updates the SSP's fast-switching cache.
The following example recomputes the entries in the IPX SSE fast-switching cache:
clear ipx sse
ipx route-cache
Use the clear sse privileged EXEC command to reinitialize the route processor on the Cisco 7000 series.
clear sseThis command has no arguments or keywords.
Disabled
Privileged EXEC
The silicon switching engine (SSE) is on the Silicon Switch Processor (SSP) board in the
Cisco 7000.
The following example causes the route processor to be reinitialized:
clear sse
To apply a generic output filter to an interface, use ipx access-group interface configuration command. To remove the access list, use the no form of this command.
ipx access-group access-list-number| access-list-number | Number of the access list. All outgoing packets defined with either standard or extended access lists and forwarded through the interface are filtered by the entries in this access list. For standard access lists, access-list-number is a decimal number from 800 to 899. For extended access lists, access-list-number is a decimal number from 900 to 999. |
No filters are predefined.
Interface configuration
Generic filters control which packets are sent out an interface based on the packet's source and destination addresses, IPX protocol type, and source and destination socket numbers. You use the standard access-list and extended access-list commands to specify the filtering conditions.
You can apply only one generic filter to an interface.
In the following example, access list 801 is applied to Ethernet interface 1:
interface ethernet 1 ipx access-group 801
A dagger (+) indicates that the command is documented in another chapter.
access-list (standard)
access-list (extended)
priority-list protocol +
To enable IPX accounting, use the ipx accounting interface configuration command. To disable IPX accounting, use the no form of this command.
ipx accountingThis command has no arguments or keywords.
Disabled
Interface configuration
IPX accounting allows you to collect information about IPX packets and the number of bytes that are switched through the router. You collect information based on the source and destination IPX address. Accounting tracks only IPX traffic that is passing out of the router; it does not track traffic generated by or terminating at the router.
IPX accounting statistics will be accurate even if IPX fast switching is enabled or if IPX access lists are being used. However, IPX accounting does not keep statistics if autonomous switching is enabled.
The router software maintains two accounting databases, an active database and a checkpointed database.
The following example enables IPX accounting on Ethernet interface 0:
interface ethernet 0 ipx accounting
clear ipx accounting
ipx accounting-list
ipx accounting-threshold
ipx accounting-transits
show ipx accounting
To filter the networks for which IPX accounting information is kept, use the ipx accounting-list global configuration command. To remove the filter, use the no form of this command.
ipx accounting-list number mask| number | Network number. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE.
You do not need to specify leading zeros in the network number. For example, for the network number 000000AA you can enter AA. |
| mask | Network mask. |
No filters are predefined.
Global configuration
The source and destination addresses of each IPX packet are logically ANDed with the mask and compared with the network number. If there is a match, accounting information about the IPX packet is entered into the accounting database. If there is no match, the IPX packet is considered to be a transit packet and may be counted, depending on the setting of the ipx accounting-transits global configuration command.
The following example adds all networks with IPX network numbers beginning with 1 to the list of networks for which accounting information is kept:
ipx accounting-list 1 0000.0000.0000
clear ipx accounting
ipx accounting
ipx accounting-threshold
ipx accounting-transits
show ipx accounting
To set the maximum number of accounting database entries, use the ipx accounting-threshold global configuration command. To restore the default, use the no form of this command.
ipx accounting-threshold threshold| threshold | Maximum number of entries (source and destination address pairs) that the router can accumulate. |
512 entries
Global configuration
The accounting threshold defines the maximum number of entries (source and destination address pairs) that the router accumulates. The threshold is designed to prevent IPX accounting from consuming all available free memory. This level of memory consumption could occur in a router that is switching traffic for many hosts. To determine whether overflows have occurred, used the show ipx accounting EXEC command.
The following example sets the IPX accounting database threshold to 500 entries:
ipx accounting-threshold 500
clear ipx accounting
ipx accounting
ipx accounting-list
ipx accounting-transits
show ipx accounting
To set the maximum number of transit entries that will be stored in the IPX accounting database, use the ipx accounting-transits global configuration command. To disable this function, use the no form of this command.
ipx accounting-transits count| count | Number of transit entries that will be stored in the IPX accounting database. |
0 entries
Global configuration
Transit entries are those that do not match any of the filters specified by ipx accounting-list global configuration commands. If you have not defined any filters, no transit entries are possible.
To maintain accurate accounting totals, the router software maintains two accounting databases: an active database and a checkpointed database.
The following example specifies a maximum of 100 transit records to be stored in the IPX accounting database:
ipx accounting-transits 100
clear ipx accounting
ipx accounting
ipx accounting-list
ipx accounting-threshold
show ipx accounting
To set the default interpacket delay for RIP updates sent on all interfaces, use the ipx default-output-rip-delay global configuration command. To return to the initial default delay value, use the no form of this command.
ipx default-output-rip-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet RIP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between routing update packets).
Global configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet routing update. The ipx default-output-rip-delay command sets a default interpacket delay for all interfaces.
The system uses the delay specified by the ipx default-output-rip-delay command for periodic and triggered routing updates when no delay is set for periodic and triggered routing updates on an interface. When you set a delay for triggered routing updates, the system uses the delay specified by the ipx default-output-rip-delay command for only the periodic routing updates sent on all interfaces.
To set a delay for triggered routing updates, see the ipx triggered-rip-delay or ipx default-triggered-rip-delay commands.
Novell recommends a delay of 55 ms for compatibility with older and slower IPX machines. These machines may lose RIP updates because they process packets more slowly than the router sends them. The delay imposed by this command forces the router to pace its output to the slower-processing needs of these IPX machines.
The default delay on a NetWare 3.11 server is about 100 ms.
This command is also useful on limited bandwidth point-to-point links or X.25 and Frame Relay multipoint interfaces.
The following example sets a default interpacket delay of 55 ms for RIP updates sent on all interfaces:
ipx default-output-rip-delay 55
ipx default-triggered-rip-delay
ipx output-rip-delay
ipx triggered-rip-delay
To set a default interpacket delay for SAP updates sent on all interfaces, use the ipx default-output-sap-delay global configuration command. To return to the initial default delay value, use the no form of this command.
ipx default-output-sap-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet SAP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between update packets).
Global configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet SAP update. The ipx default-output-sap-delay command sets a default interpacket delay for all interfaces.
The system uses the delay specified by the ipx default-output-sap-delay command for periodic and triggered SAP updates when no delay is set for periodic and triggered updates on an interface. When you set a delay for triggered updates, the system uses the delay specified by the ipx default-output-sap-delay command only for the periodic SAP updates sent on all interfaces.
To set a delay for triggered updates, see the ipx triggered-sap-delay or ipx default-triggered-sap-delay commands.
Novell recommends a delay of 55 ms for compatibility with older and slower IPX servers. These servers may lose SAP updates because they process packets more slowly than the router sends them. The delay imposed by this command forces the router to pace its output to the slower-processing needs of these servers.
The default delay on a NetWare 3.11 server is about 100 ms.
This command is also useful on limited bandwidth point-to-point links or X.25 interfaces.
The following example sets a default interpacket delay of 55 ms for SAP updates sent on all interfaces:
ipx default-output-sap-delay 55
ipx default-triggered-sap-delay
ipx output-sap-delay
ipx triggered-sap-delay
To set the default interpacket delay for triggered RIP updates sent on all interfaces, use the ipx default-triggered-rip-delay global configuration command. To return to the system default delay, use the no form of this command.
ipx default-triggered-rip-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet RIP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between routing update packets).
Global configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet routing update. A triggered routing update is one that the system sends in response to a "trigger" event, such as a request packet, interface up/down, route up/down, or server up/down.
The ipx default-triggered-rip-delay command sets the default interpacket delay for triggered routing updates sent on all interfaces. On a single interface, you can override this global default delay for triggered routing updates using the ipx triggered-rip-delay interface command.
The global default delay for triggered routing updates overrides the delay value set by the ipx output-rip-delay or ipx default-output-rip-delay command for triggered routing updates.
If the delay value set by the ipx output-rip-delay or ipx default-output-rip-delay command is high, then we strongly recommend a low delay value for triggered routing updates so that updates triggered by special events are sent in a more timely manner than periodic routing updates.
Novell recommends a delay of 55 ms for compatibility with older and slower IPX machines. These machines may lose RIP updates because they process packets more slowly than the router sends them. The delay imposed by this command forces the router to pace its output to the slower-processing needs of these IPX machines.
The default delay on a NetWare 3.11 server is about 100 ms.
When you do not set the interpacket delay for triggered routing updates, the system uses the delay specified by the ipx output-rip-delay or ipx default-output-rip-delay command for both periodic and triggered routing updates.
When you use the no form of the ipx default-triggered-rip-delay command, the system uses the delay set by the ipx output-rip-delay or ipx default-output-rip-delay command for triggered RIP updates, if set. Otherwise, the system uses the initial default delay as described in the "Default" section.
This command is also useful on limited bandwidth point-to-point links or X.25 and Frame Relay multipoint interfaces.
The following example sets an interpacket delay of 55 ms for triggered routing updates sent on all interfaces:
ipx default-triggered-rip-delay 55
ipx default-output-rip-delay
ipx output-rip-delay
ipx triggered-rip-delay
To set the default interpacket delay for triggered SAP updates sent on all interfaces, use the ipx default-triggered-sap-delay global configuration command. To return to the system default delay, use the no form of this command.
ipx default-triggered-sap-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet SAP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between update packets).
Global configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet SAP update. A triggered SAP update is one that the system sends in response to a "trigger" event, such as a request packet, interface up/down, route up/down, or server up/down.
The ipx default-triggered-sap-delay command sets the default interpacket delay for triggered SAP updates sent on all interfaces. On a single interface, you can override this global default delay for triggered updates using the ipx triggered-sap-delay interface command.
The global default delay for triggered updates overrides the delay value set by the ipx output-sap-delay or ipx default-output-sap-delay command for triggered updates.
If the delay value set by the ipx output-sap-delay or ipx default-output-sap-delay command is high, then we strongly recommend a low delay value for triggered updates so that updates triggered by special events are sent in a more timely manner than periodic updates.
Novell recommends a delay of 55 ms for compatibility with older and slower IPX servers. These servers may lose SAP updates because they process packets more slowly than the router sends them. The delay imposed by this command forces the router to pace its output to the slower-processing needs of these IPX servers.
The default delay on a NetWare 3.11 server is about 100 ms.
When you do not set the interpacket delay for triggered SAP updates, the system uses the delay specified by the ipx output-sap-delay or ipx default-output-sap-delay command for both periodic and triggered SAP updates.
When you use the no form of the ipx default-triggered-sap-delay command, the system uses the delay set by the ipx output-sap-delay or ipx default-output-sap-delay command for triggered SAP updates, if set. Otherwise, the system uses the initial default delay as described in the "Default" section.
This command is also useful on limited bandwidth point-to-point links or X.25 and Frame Relay multipoint interfaces.
The following example sets an interpacket delay of 55 ms for triggered SAP updates sent on all interfaces:
ipx default-triggered-sap-delay 55
ipx default-output-sap-delay
ipx output-sap-delay
ipx triggered-sap-delay
To set the tick count, use the ipx delay interface configuration command. To reset the default increment in the delay field, use the no form of this command.
ipx delay number| number | Number of IBM clock ticks of delay to use. One clock tick is 1/18th of a second (approximately 55 milliseconds). |
Determined from the delay configured on the interface with the delay command. It is (the interface delay + 333) / 334. Therefore, unless you change the delay by a value greater than 334, you will not notice a difference.
Interface configuration
The ipx delay command sets the count used in the IPX RIP delay field, which is also known as the ticks field.
Leaving the delay at its default value is sufficient for most interfaces.
If the link is an IPXWAN link, it determines its delay dynamically and the ipx delay command has no effect.
The following example changes the delay for serial interface 0 to 10 ticks:
interface serial 0 ipx delay 10
A dagger (+) indicates that the command is documented in another chapter.
delay +
ipx maximum-paths
ipx output-network-filter
ipx output-rip-delay
To administratively shut down an IPX network, use the ipx down interface configuration command. To restart the network, use the no form of this command.
ipx down network| network | Number of the network to shut down. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
Disabled
Interface configuration
The ipx down command administratively shuts down the specified network. 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. This allows the neighboring systems to update their routing, SAP, and other tables without having to wait for routes and services learned via this network to time out.
The following example administratively shuts down network AA on Ethernet interface 0:
interface ethernet 0 ipx down AA
To change the delay when responding to Get Nearest Server (GNS) requests, use the ipx gns-response-delay global configuration command. To return to the default delay, use the no form of this command.
ipx gns-response-delay [milliseconds]| milliseconds | (Optional) Time, in milliseconds, that the router waits after receiving a Get Nearest Server request from an IPX client before responding with a server name to that client. The default time is 0 milliseconds. A value of zero indicates no delay. |
0 milliseconds
Global configuration
The delay in responding to Get Nearest Server requests is imposed so that in certain topologies any local Novell IPX servers respond to the GNS requests before our router does. It is desirable to have these end-host server systems get their reply to the client before the router does, because the client typically takes the first response, not the best, and in this case the best response is the one from the local server.
In situations in which servers are always located across routers from their clients, there is no need for a delay to be imposed.
The following example eliminates delay when responding to GNS requests:
ipx gns-response-delay 0
To rotate using a round-robin selection method through a set of eligible servers when responding to Get Nearest Server (GNS) requests, use the ipx gns-round-robin global configuration command. To use the most recently learned server, use the no form of this command.
ipx gns-round-robinThe command has no arguments or keywords.
The most recently learned, eligible server is used.
Global configuration
In the normal server selection process, requests for service are responded to with the most recently learned, closest server. If you enable the round-robin method, the router maintains a list of the nearest servers eligible to provide specific services. It uses this list when responding to Get Nearest Server (GNS) requests. Responses to requests are distributed in a round-robin fashion across all active IPX interfaces on the router.
Eligible servers are those that satisfy the "nearest" requirement for a given request and that are not filtered either by a SAP filter or by a GNS filter.
The following example responds to GNS requests using a round-robin selection method from a list of eligible nearest servers:
ipx gns-round-robin
ipx output-gns-filter
ipx output-sap-filter
To forward broadcast packets (except type 20 propagation packets) to a specified server, use the ipx helper-address interface configuration command. To disable this function, use the no form of this command.
ipx helper-address network.node| network | Network on which the target IPX server resides. This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. A network number of -1 indicates all-nets flooding. You do not need to specify leading zeros in the network number. For example, for the network number 000000AA you can enter just AA. |
| node | Node number of the target Novell server. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). A node number of FFFF.FFFF.FFFF matches all servers. |
Disabled
Interface configuration
Routers normally block all broadcast requests and do not forward them to other network segments. This is done to prevent the degradation of performance over the entire network. The ipx helper-address command allows broadcasts to be forwarded to other networks (except type 20 propagation packets). This is useful when a network segment does not have an end-host capable of servicing a particular type of broadcast request. This command lets you forward the broadcasts to a server, network, or networks that can process them. Incoming unrecognized broadcast packets that match the access list created with the ipx helper-list command, if it is present, are forwarded.
Note that type 20 propagation packet handling is controlled by a separate mechanism. See the discussion of the ipx type-20-packet-propagation command for more information.
You can specify multiple ipx helper-address commands on a given interface.
Our routers support all-networks flooded broadcasts (sometimes referred to as all-nets flooding). These are broadcast messages that are forwarded to all networks. To configure the all-nets flooding, define the IPX helper address for an interface as follows:
ipx helper-address -1.FFFF.FFFF.FFFF
On systems configured for IPX routing, this helper address is displayed as follows (via the show ipx interface command):
FFFFFFFF.FFFF.FFFF.FFFF
Although our routers take care to keep broadcast traffic to a minimum, some duplication is unavoidable. When loops exist, all-nets flooding can propagate bursts of excess traffic that will eventually age out when the hop count reaches its limit (16 hops). Use all-nets flooding carefully and only when necessary. Note that you can apply additional restrictions by defining a helper list.
In the following example, all-nets broadcasts on Ethernet interface 0 (except type 20 propagation packets) are forwarded to IPX server 00b4.23cd.110a on network bb:
interface ethernet 0 ipx helper-address bb.00b4.23cd.110a
ipx helper-list
ipx type-20-propagation
To assign an access list to an interface to control broadcast traffic (including type 20 propagation packets), use the ipx helper-list interface configuration command. To remove the access list from an interface, use the no form of this command.
ipx helper-list access-list-number| access-list-number | Number of the access list. All outgoing packets defined with either standard or extended access lists are filtered by the entries in this access list. For standard access lists, access-list-number is a decimal number from 800 to 899. For extended access lists, it is a decimal number from 900 to 999. |
No access list is preassigned.
Interface configuration
The ipx helper-list command specifies an access list to use in forwarding broadcast packets. One use of this command is to prevent client nodes from discovering services they should not use.
Because the destination address of a broadcast packet is by definition the broadcast address, this command is useful only for filtering based on the source address of the broadcast packet.
The helper list, if present, is applied to both all-nets broadcast packets and type 20 propagation packets.
The helper list on the input interface is applied to packets before they are output via either the helper address or type 20 propagation packet mechanism.
You should filter IPX broadcasts on dial-on-demand routing (DDR) and other similar interfaces, because IPX sends broadcast messages very regularly.
The following example assigns access list 900 to Ethernet interface 0 to control broadcast traffic:
interface ethernet 0 ipx helper-list 900
access-list (standard)
access-list (extended)
ipx helper-address
ipx type-20-propagation
To control which networks are added to the router's routing table, use the ipx input-network-filter interface configuration command. To remove the filter from the interface, use the no form of this command.
ipx input-network-filter access-list-number| access-list-number | Number of the access list. All incoming packets defined with either standard or extended access lists are filtered by the entries in this access list. For standard access lists, access-list-number is a decimal number from 800 to 899. For extended access lists, it is a decimal number from 900 to 999. |
No filters are predefined.
Interface configuration
The ipx input-network-filter command controls which networks are added to the routing table based on the networks learned in incoming IPX routing updates (RIP updates) on the interface.
You can issue only one ipx input-network-filter command on each interface.
In the following example, access list 876 controls which networks are added to the routing table when IPX routing updates are received on Ethernet interface 1. Routing updates for network 1b will be accepted. Routing updates for all other networks are implicitly denied and are not added to the routing table.
access-list 876 permit 1b interface ethernet 1 ipx input-network-filter 876
The following example is a variation of the preceding that explicitly denies network 1a and explicitly allows updates for all other networks:
access-list 876 deny 1a access-list 876 permit -1
access-list (standard)
access-list (extended)
ipx output-network-filter
ipx router-filter
To control which services are added to the router's SAP table, use the ipx input-sap-filter interface configuration command. To remove the filter, use the no form of this command.
ipx input-sap-filter access-list-number| access-list-number | Number of the SAP access list. All incoming packets are filtered by the entries in this access list. The argument access-list-number is a decimal number from 1000 to 1099. |
No filters are predefined.
Interface configuration
The ipx input-sap-filter command filters all incoming service advertisements received by the router. This is done prior to a router's accepting information about a service.
You can issue only one ipx input-sap-filter command on each interface.
When configuring SAP filters for NetWare 3.11 and later servers, use the server's internal network and node number (the node number is always 0000.0000.0001) as its address in the SAP access-list command. Do not use the network.node address of the particular interface board.
The following example denies service advertisements about the server at address 3c.0800.89a1.1527, but accepts information about all other services on all other networks:
access-list 1000 deny 3c.0800.89a1.1527 access-list 1000 permit -1 interface ethernet 0 ipx input-sap-filter 1000
access-list (SAP filtering)
ipx output-sap-filter
ipx router-sap-filter
To configure the IPXWAN protocol on a serial interface, use the ipx ipxwan interface configuration command. To disable the IPXWAN protocol, use the no form of this command.
ipx ipxwan local-node number local-server [retry-interval] [retry-attempts]| local-node | Primary network number of the router. This is an IPX network number that is unique across the entire internet. On NetWare 3.x servers, the primary network number is called the internal network number. The router with the higher number is determined to be the link Master. |
| number | The IPX network number to be used for the link if this router is the one determined to be the link master. The number is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE.
You do not need to specify leading zeros in the network number, e.g., for the network number 000000AA, enter AA. |
| local-server | Name of the local router. It can be up to 47 characters long, and can contain uppercase letters, digits, underscores (_), hyphens (-), and at signs (@). On NetWare 3.x servers, the local server is called the router name. |
| retry-interval | (Optional) Retry interval, in seconds. This interval defines how often the router will retry failed negotiation attempts. It can be a value from 1 through 600. The default is 20 seconds. |
| retry-attempts | (Optional) Maximum number of times the router will retry failed negotiation attempts. It can be a value from 1 through 100. The default is 3. The router intentionally ignores the IPXWAN retry counter. The router continues to send out TIMER_REQUEST packets until it receives a TIMER_RESPONSE packet. |
IPXWAN is disabled.
retry-interval: 20 seconds
retry-attempts: 3
Interface configuration
IPXWAN is a startup end-to-end options negotiations protocol. When a link comes up, the first IPX packets sent across are IPXWAN packets negotiating the options for the link. When the IPXWAN options have been successfully determined, then normal IPX traffic starts. The three options negotiated are the link IPX network number, Ethernet network number, and link delay (ticks) characteristics.The side of the link with the higher local-node number (internal network number) gives the IPX network number and delay to use for the link to the other side. Once IPXWAN finishes, no IPXWAN packets are sent unless link characteristics change or the connection fails. For example, if the IPX delay is changed from the default setting, an IPXWAN restart will be forced.
To enable the IPXWAN protocol on a serial interface, you must not have configured an IPX network number (using the ipx network interface configuration command) on that interface.
If you want to control the delay on a link, use the ipx delay interface configuration command. If you issue this command when the serial link is already up, the state of the link will be reset and renegotiated.
The following example enables IPXWAN on serial interface 1 on device "CHICAGO-RTR":
interface serial1 no ipx network encapsulation ppp ipx ipxwan 6666 100 CHICAGO-RTR
On the remote side:
interface serial0 no ipx network encapsulation ppp ipx ipxwan 1000 101 NYC-RTR
When the link comes up, CHICAGO-RTR will be the "Master" because of its higher internal network number. It will give the IPX number 100 to NYC-RTR to use as the network number for the link. The link delay (in ticks) will be determined by the exchange of packets between the two routers.
A dagger (+) indicates that the command is documented in another chapter.
encapsulation ppp +
ipx delay
ipx network
show ipx interfaces
To set the maximum number of equal-cost paths the router uses when forwarding packets, use the ipx maximum-paths global configuration command. To restore the default value of 1, use the no form of this command.
ipx maximum-paths paths| paths | Maximum number of equal-cost paths which the router will use. The argument paths can be a value from 1 to 512. The default value is 1. |
1 path
Global configuration
The ipx maximum-paths command is designed to increase throughput by allowing the router to choose among several equal-cost, parallel paths. (Note that when paths have differing costs, the router chooses lower-cost routes in preference to higher-cost routes.) IPX does load sharing on a packet-by-packet basis in round-robin fashion, regardless of whether you are using fast switching or process switching. 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.
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.
In the following example, the router uses up to three parallel paths:
ipx maximum-paths 3
ipx delay
show ipx route
To control incoming IPX NetBIOS messages, use the ipx netbios input-access-filter interface configuration command. To remove the filter, use the no form of this command.
ipx netbios input-access-filter {host | bytes} name| host | Indicates that the following argument is the name of a NetBIOS access filter previously defined with one or more netbios access-list host commands. |
| bytes | Indicates that the following argument is the name of a NetBIOS access filter previously defined with one or more netbios access-list bytes commands. |
| name | Name of a NetBIOS access list. |
No filters are predefined.
Interface configuration
You can issue only one ipx netbios input-access-filter host and one ipx netbios input-access-filter bytes command on each interface.
These filters apply only to IPX NetBIOS packets. They have no effect on LLC2 NetBIOS packets.
The following example filters packets arriving on Token Ring interface 1 using the NetBIOS access list "engineering":
netbios access-list host engineering permit eng* netbios access-list host engineering deny manu* interface token 1 ipx netbios input-access-filter engineering
ipx netbios output-access filter
netbios access-list
show ipx interface
To control outgoing NetBIOS messages, use the ipx netbios output-access-filter interface configuration command. To remove the filter, use the no form of this command.
ipx netbios output-access-filter {host | bytes} name| host | Indicates that the following argument is the name of a NetBIOS access filter previously defined with one or more netbios access-list host commands. |
| bytes | Indicates that the following argument is the name of a NetBIOS access filter previously defined with one or more netbios access-list bytes commands. |
| name | Name of a previously defined NetBIOS access list. |
No filters are predefined.
Interface configuration
You can issue only one ipx netbios output-access-filter host and one ipx netbios output-access-filter bytes command on each interface.
These filters apply only to IPX NetBIOS packets. They have no effect on LLC2 NetBIOS packets.
The following example filters packets leaving Token Ring interface 1 using the NetBIOS access list "engineering":
netbios access-list bytes engineering permit 20 AA**04 interface token 1 ipx netbios output-access-filter bytes engineering
ipx netbios input-access filter
netbios access-list
show ipx interface
To enable IPX routing on a particular interface and to optionally select the type of encapsulation (framing), use the ipx network interface configuration command. To disable IPX routing, use the no form of this command.
ipx network number [encapsulation encapsulation-type [secondary]]IPX routing is disabled.
Encapsulation types:
For Ethernet: novell-ether
For Token Ring: sap
For FDDI: snap
Interface configuration
The extended ipx network command allows you to configure more than one logical network on the same physical network (network cable segment). Each network on a given interface must have a different encapsulation type. The first network you configure on an interface is considered to be the primary network. Any additional networks are considered to be secondary networks; these must include the secondary keyword. You also can use this command to configure a single logical network on a physical network.
You can configure an IPX network on any supported interface as long as all the networks on the same physical interface use a distinct encapsulation type. For example, you can configure up to four IPX networks on a single Ethernet cable because Ethernet supports four encapsulation types.
The interface processes only packets with the correct encapsulation and the correct network number. IPX networks using other encapsulations can be present on the physical network. The only effect on the router is that it uses some processing time to examine packets to determine whether they have the correct encapsulation.
If you use the standard ipx network and ipx encapsulation commands on an interface that supports more than one logical network, you can use them only to define the first, or primary, network on the interface. Use the extended ipx network command to define additional logical networks.
All logical networks on an interface share the same set of configuration parameters. For example, if you change the IPX RIP update time on an interface, you change it for all networks on that interface.
This command is useful when migrating from one type of encapsulation to another. If you are using it for this purpose, you should define the new encapsulation on the primary network.
To delete all networks on an interface, use the following command:
no ipx networkDeleting the primary network with the following command also deletes all networks on that interface. The argument number is the number of the primary network.
no ipx network numberTo delete a secondary network on an interface, use one of the following commands. The argument number is the number of a secondary network.
no ipx network numberThe following two commands also allow you to enable IPX routing on an interface and specify the encapsulation. These commands can be used on interfaces that support a single network or when enabling the primary subinterface on an interface that supports multiple networks. They are supported in this release of the software to provide compatibility with older versions of the router software.
ipx networkThe following example configures an interface that has four logical networks:
interface ethernet 0 ipx network 0123 ipx encapsulation snap ipx network 0234 encapsulation sap secondary ipx network 0345 encapsulation arpa secondary ipx network 0456 encapsulation novell-ether secondary
ipx routing
To control which servers are included in the Get Nearest Server (GNS) responses sent by the router, use the ipx output-gns-filter interface configuration command. To remove the filter from the interface, use the no form of this command.
ipx output-gns-filter access-list-number| access-list-number | Number of the SAP access list. All outgoing GNS packets are filtered by the entries in this access list. The argument access-list-number is a decimal number from 1000 to 1099. |
No filters are predefined.
Interface configuration
You can issue only one ipx output-gns-filter command on each interface.
The following example excludes the server at address 3c.0800.89a1.1527 from GNS responses sent on Ethernet interface 0, but allows all other servers:
access-list 1000 deny 3c.0800.89a1.1527 access-list 1000 permit -1 ipx routing interface ethernet 0 ipx output-gns-filter 1000
access-list (SAP filtering)
ipx gns-round-robin
To control the list of networks included in routing updates sent out an interface, use the ipx output-network-filter interface configuration command. To remove the filter from the interface, use the no form of this command.
ipx output-network-filter access-list-number| access-list-number | Number of the access list. All outgoing packets defined with either standard or extended access lists are filtered by the entries in this access list. For standard access lists, access-list-number is a decimal number from 800 to 899. For extended access lists, it is a decimal number from 900 to 999. |
No filters are predefined.
Interface configuration
The ipx output-network-filter command controls which networks the router advertises in its IPX routing updates (RIP updates).
You can issue only one ipx output-network-filter command on each interface.
In the following example, access list 896 controls which networks are specified in routing updates sent out the serial 1 interface. This configuration causes network 2b to be the only network advertised in Novell routing updates sent on the specified serial interface.
access-list 896 permit 2b interface serial 1 ipx output-network-filter 896
access-list (standard)
access-list (extended)
ipx input-network-filter
ipx router-filter
To set the interpacket delay for RIP updates sent on a single interface, use the ipx output-rip-delay interface configuration command. To return to the default value, use the no form of this command.
ipx output-rip-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet RIP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between routing update packets).
Interface configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet routing update. The ipx output-rip-delay command sets the interpacket delay for a single interface.
The system uses the interpacket delay specified by the ipx output-rip-delay command for periodic and triggered routing updates when no delay is set for triggered routing updates. When you set a delay for triggered routing updates, the system uses the delay specified by the ipx output-rip-delay command for only the periodic routing updates sent on the interface.
To set a delay for triggered routing updates, see the ipx triggered-rip-delay or ipx default-triggered-rip-delay commands.
You can also set a default RIP interpacket delay for all interfaces. See the ipx default-output-rip-delay command for more information.
Novell recommends a delay of 55 ms for compatibility with older and slower IPX machines. These machines may lose RIP updates because they process packets more slowly than the router sends them. The delay imposed by this command forces the router to pace its output to the slower-processing needs of these IPX machines.
The default delay on a NetWare 3.11 server is about 100 ms.
This command is also useful on limited bandwidth point-to-point links or X.25 and Frame Relay multipoint interfaces.
The following example establishes a 55-ms interpacket delay on serial interface 0:
interface serial 0 ipx network 106A ipx output-rip-delay 55
ipx update-time
ipx default-output-rip-delay
ipx default-triggered-rip-delay
ipx triggered-rip-delay
To set the interpacket delay for SAP updates sent on a single interface, use the ipx output-sap-delay interface configuration command. To return to the default delay value, use the no form of this command.
ipx output-sap-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet SAP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between update packets).
Interface configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet SAP update. The ipx output-sap-delay command sets the interpacket delay for a single interface.
The system uses the interpacket delay specified by the ipx output-sap-delay command for periodic and triggered SAP updates when no delay is set for triggered updates. When you set a delay for triggered updates, the system uses the delay specified by the ipx output-sap-delay command only for the periodic updates sent on the interface.
To set a delay for triggered updates, see the ipx triggered-sap-delay or ipx default-triggered-sap-delay commands.
You can also set a default SAP interpacket delay for all interfaces. See the ipx default-output-sap-delay command for more information.
Novell recommends a delay of 55 ms for compatibility with older and slower IPX servers. These servers may lose SAP updates because they process packets more slowly than the router sends them. The delay imposed by the ipx output-sap-delay command forces the router to pace its output to the slower-processing needs of these servers.
The default delay on a NetWare 3.11 server is about 100 ms.
This command is also useful on limited bandwidth point-to-point links or X.25 and Frame Relay multipoint interfaces.
The following example establishes a 55-ms delay between packets in multiple-packet SAP updates on Ethernet interface 0:
interface ethernet 0 ipx network 106A ipx output-sap-delay 55
ipx default-output-sap-delay
ipx default-triggered-sap-delay
ipx sap-interval
ipx triggered-sap-delay
To control which services are included in Service Advertisement Protocol (SAP) updates sent by the router, use the ipx output-network-filter interface configuration command. To remove the filter, use the no form of this command.
ipx output-sap-filter access-list-number| access-list-number | Number of the SAP access list. All outgoing service advertisements are filtered by the entries in this access list. The variable access-list-number is a decimal number from 1000 to 1099. |
No filters are predefined.
Interface configuration
The router applies output SAP filters prior to sending SAP packets.
You can issue only one ipx output-sap-filter command on each interface.
When configuring SAP filters for NetWare 3.11 and later servers, use the server's internal network and node number (the node number is always 0000.0000.0001) as its address in the SAP access-list command. Do not use the network.node address of the particular interface board.
The following example denies service advertisements about server 0000.0000.0001 on network aa from being send on network 4d (via Ethernet interface 1). All other services are advertised via this network. All services, included those from server aa.0000.0000.0001, are advertised via networks 3c and 2b.
access-list 1000 deny aa.0000.0000.0001 access-list 1000 permit -1 interface ethernet 0 ipx net 3c interface ethernet 1 ipx network 4d ipx output-sap-filter 1000 interface serial 0 ipx network 2b
access list (SAP filtering)
ipx gns-round-robin
ipx input-sap-filter
ipx router-sap-filter
To control whether odd-length packets are padded so as to be sent as even-length packets on an interface, use the ipx pad-process-switched-packets interface configuration command. To disable padding, use the no form of this command.
ipx pad-process-switched-packetsThis command has no arguments or keywords.
Enabled on Ethernet interfaces
Disabled on Token Ring, FDDI, and serial interfaces.
Interface configuration
Use this command only under the guidance of a customer engineer or other service representative.
The ipx pad-process-switched-packets command affects process-switched packets only, so you must disable fast switching before the ipx pad-process-switched-packets command has any effect.
Some IPX end hosts reject Ethernet packets that are not padded. Certain topologies can result in such packets being forwarded onto a remote Ethernet network. Under specific conditions, padding on intermediate media can be used as a temporary workaround for this problem.
ipx route-cache
To add a static route to the routing table, use the ipx route global configuration command. To remove a route from the routing table, use the no form of this command.
ipx route network network.node| network | Network to which you want to establish a static route.
This is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. |
| network.node | Router to which to forward packets destined for the specified network.
The argument network is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. You do not need to specify leading zeros in the network number. For example, for the network number 000000AA, you can enter just AA. The argument node is the node number of the target router. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
No static routes are predefined.
Global configuration
The ipx route command forwards packets destined for the specified network (network) via the specified router (network.node), regardless of whether that router is sending dynamic routing information.
Be careful when assigning static routes. When links associated with static routes are lost, traffic may stop being forwarded, even though alternative paths might be available.
In the following example, the router at address 3abc.0000.0c00.1ac9 handles all traffic destined for network 5e:
ipx routing ipx route 5e 3abc.0000.0c00.1ac9
show ipx route
To enable IPX fast switching and autonomous switching, use the ipx route-cache interface configuration command. To disable fast switching, use the no form of this command.
ipx route-cache [cbus | sse]| no arguments | Enables fast switching. |
| cbus | (Optional) Enables IPX autonomous switching. |
| sse | (Optional) Enables SSE fast switching. |
Fast switching enabled
Autonomous switching disabled
SSE switching disabled
Interface configuration
Fast switching allows higher throughput by switching packets using a cache created by previous transit packets. On ciscoBus-2 interface cards, fast switching is done between all encapsulation types. On other interface cards, fast switching is done in all cases except the following: transfer of packets with sap encapsulation from an Ethernet, a Token Ring, or an FDDI network to a standard serial line.
You might want to disable fast switching in two situations. One is if you want to save memory on the interface cards: fast-switching caches require more memory than those used for standard switching. The second situation is to avoid congestion on interface cards when a high-bandwidth interface is writing large amounts of information to a low-bandwidth interface.
Autonomous switching provides faster packet switching by allowing the ciscoBus processor to switch packets independently without having to interrupt the system processor. It is available only in Cisco 7000 systems, and in AGS+ systems with high-speed network controller CBus2-only interfaces, such as the CCTL2 ciscoBus controller running microcode version 11.0 or later.
SSE fast switching uses the silicon switching engine (SSE) on the Cisco 7000 Series SSP card to perform packet switching.
The following example enables fast switching and autonomous switching on an interface:
interface ethernet 0 ipx route-cache cbus
The following example enables fast switching and SSE fast switching on an interface:
interface ethernet 0/1 ipx route-cache sse
In the following example, both fast switching and autonomous switching are turned off on an interface:
interface ethernet 0 no ipx route-cache
Assuming that Ethernet 0 has ipx route-cache and ipx route-cache cbus is enabled, the following example turns off only autonomous switching on an interface, but leaves fast switching enabled:
interface ethernet 0 no ipx route-cache cbus
clear ipx cache
ipx source-network-update
ipx watchdog-spoof
show ipx cache
To control the routers from which packets are accepted, use the ipx router-filter interface configuration command. To remove the filter from the interface, use the no form of this command.
ipx router-filter access-list-number| access-list-number | Number of the access list. All incoming packets defined with either standard or extended access lists are filtered by the entries in this access list. For standard access lists, access-list-number is a decimal number from 800 to 899. For extended access lists, it is a decimal number from 900 to 999. |
No filters are predefined.
Interface configuration
You can issue only one ipx router-filter command on each interface.
In the following example, access list 866 controls the routers from which packets are accepted. For Ethernet interface 0, only packets from the router at 3c.0000.00c0.047d are accepted. All other packets are implicitly denied.
access-list 866 permit 3c.0000.00c0.047d interface ethernet 0 ipx router-filter 866
access-list (standard)
access-list (extended)
ipx input-network-filter
ipx output-network-filter
To filter Service Advertisement Point (SAP) messages received from a particular router, use the ipx router-sap-filter interface configuration command. To remove the filter, use the no form of this command.
ipx router-sap-filter access-list-number| access-list-number | Number of the access list. All incoming service advertisements are filtered by the entries in this access list. The argument access-list-number is a decimal number from 1000 to 1099. |
No filters are predefined.
Interface configuration
You can issue only one ipx router-sap-filter command on each interface.
In the following example, the router will receive service advertisements only from router aa.0207.0104.0874:
access-list 1000 permit aa.0207.0104.0874 access-list 1000 deny -1 interface ethernet 0 ipx router-sap-filter 1000
access-list (SAP filtering)
ipx input-sap-filter
ipx output-sap-filter
ipx sap
show ipx interface
To enable IPX routing, use the ipx routing global configuration command. To disable IPX routing, use the no form of this command.
ipx routing [node]| node | (Optional) Node number of the router. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). It must not be a multicast address.
If you omit node, the router 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 no satisfactory interfaces are present in the router (such as only serial interfaces), you must specify node. |
Disabled
Global configuration
The ipx routing command enables the IPX Routing Information Protocol (RIP) and Service Advertisement Point (SAP) services on the router.
If you omit the argument node and if the MAC address later changes, the IPX node address automatically changes to the new address. However, connectivity may be lost between the time that the MAC address changes and the time that the IPX clients and servers learn the router's new address.
If you plan to use DECnet and IPX routing concurrently on the same interface, you should enable DECnet router 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.
The following example enables IPX routing:
ipx routing
ipx network
To specify static Service Advertisement Protocol (SAP) entries, use the ipx sap global configuration command. To remove static SAP entries, use the no form of this command.
ipx sap service-type name network.node socket hop-count| service-type | SAP service-type number. Table 19-3 earlier in this chapter lists some IPX SAP services. |
| name | Name of the server that provides the service. |
| network.node | Network number and node address of the server.
The argument network is an eight-digit hexadecimal number that uniquely identifies a network cable segment. It can be a number in the range 1 to FFFFFFFE. You do not need to specify leading zeros in the network number. For example, for the network number 000000AA you can enter just AA. The argument node is the node number of the target Novell server. This is a 48-bit value represented by a dotted triplet of four-digit hexadecimal numbers (xxxx.xxxx.xxxx). |
| socket | Socket number for this service. Table 19-2 earlier in this chapter lists some IPX socket numbers. |
| hop-count | Number of hops to the server. |
Disabled
Global configuration
The ipx sap command allows you to add static entries into the SAP table. Each entry has a SAP service associated with it. Static SAP assignments always override any identical entries in the SAP table that are learned dynamically, regardless of hop count. The router will not announce a static SAP entry unless it has a route to that network.
In the following example, the route to JOES_SERVER is not yet learned, so the system displays an informational message. The JOES_SERVER service will not be announced in the regular SAP updates until the router learns the route to it either by means of a RIP update from a neighbor or an ipx sap command.
ipx sap 107 MAILSERV 160.0000.0c01.2b72 8104 1 ipx sap 4 FILESERV 165.0000.0c01.3d1b 451 1 ipx sap 143 JOES_SERVER A1.0000.0c01.1234 8170 2 no route to A1, JOES_SERVER won't be announced until route is learned
ipx input-sap-filter
ipx output-sap-filter
ipx router-sap-filter
show ipx servers
To configure less frequent Service Advertisement Point (SAP) updates over slow links, use the ipx sap-interval interface configuration command. To return to the default value, use the no form of this command.
ipx sap-interval interval| interval | Interval, in minutes, between SAP updates sent by the router. The default value is 1 minute. If interval is 0, periodic updates are never sent. |
1 minute
Interface configuration
Setting the interval at which SAP updates are sent is most useful on limited-bandwidth, point-to-point links or on X.25 interfaces.
You should ensure that all IPX servers and routers on a given network have the same SAP interval. Otherwise, they may 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. This means that you should never change the interval for an Ethernet or Token Ring network that has servers on it.
Setting the interval to zero means that periodic SAP updates are never sent. It is recommended that you never do this. If you set the interval to zero, routers that are inaccessible for any reason when a server powers up or shuts down will miss that event, and will either fail to learn about new servers or fail to detect that the server shut down.
In the following example, SAP updates are sent (and expected) on serial interface 0 every 5 minutes:
interface serial 0 ipx sap-interval 5
ipx output-sap-delay
To configure the maximum length of the queue of pending input SAP GNS requests and SAP query packets, use the ipx sap-queue-maximum global configuration command. To return to the default value, use the no form of this command.
ipx sap-queue-maximum number| number | Maximum length of the queue of pending SAP requests. By default, there is no limit to the number of pending SAP requests that the router stores in this queue. |
No maximum queue size
Global configuration
The router 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, the router can be inundated with hundreds of requests for servers. Most of these can be repeated requests from the same clients. The ipx sap-queue-maximum command allows you to configure the maximum length allowed for the pending SAP requests queue. Packets received when the queue is full are dropped.
The following example sets the length of the queue of pending SAP requests to 20:
ipx sap-queue-maximum 20
To repair corrupted network numbers, use the ipx source-network-update interface configuration command. To disable this feature, use the no form of this command.
ipx source-network-updateThis command has no arguments or keywords.
Disabled
Interface configuration
In some early implementations of IPX client software, it was possible for the client's network number to become corrupted. The ipx source-network-update command repairs this number by setting the source network field of any packet on the local network that has a hop count of zero.
You must disable fast switching with the no ipx route-cache command before using the ipx source-network-update command.
This command interferes with the proper working of OS/2 Requestors. Therefore, do not use this command in a network that has OS/2 Requestors.
Do not use the ipx source-network-update command on interfaces on which NetWare servers are using internal network numbers.
 | Caution The ipx source-network-update command interferes with the proper working of OS/2 Requestors. Do not use this command in a network that has OS/2 Requestors. |
| Caution Do not use the ipx source-network-update command on interfaces on which NetWare (NetWare 3.1x or 4.0 or later) Servers are using internal network numbers. |
In the following example, corrupted network numbers on serial interface 0 are repaired:
interface serial 0 no ipx route-cache ipx source-network-update
ipx route-cache
To set the interpacket delay for triggered RIP updates sent on a single interface, use the ipx triggered-rip-delay interface configuration command. To return to the default delay, use the no form of this command.
ipx triggered-rip-delay delay| delay | Delay, in milliseconds, between packets in a multiple-packet RIP update. With Cisco IOS Release 10.0, the default delay is 0 ms. Novell recommends a delay of 55 ms. |
With Cisco IOS Release 10.0, the default delay is 0 ms (that is, no additional delay between routing update packets).
Interface configuration
The interpacket delay is the delay between the individual packets sent in a multiple-packet routing update. A triggered routing update is one that the system sends in response to a "trigger" event, such as a request packet, interface up/down, route up/down, or server up/down.
The ipx triggered-rip-delay command sets the interpacket delay for triggered routing updates sent on a single interface. The delay value set by this command overrides the delay value set by the ipx output-rip-delay or