3Com Router User Manual

							Configure RIP 417
RIP Version 2 does not have provisions for a zero field in its header so this 
configuration is invalid for RIP-2.
Perform the following configurations in RIP view.
Ta b l e 473   Configure Check Zero Field of RIP Version 1
RIP VERSION 1 enables zero field check by default.
Specifying the Status of 
an InterfaceYou can specify the working status of RIP on an interface, such as whether RIP is 
running on the interface and whether updated messages are transmitted or 
received on the interface.
Perform the following configurations in interface view.
Ta b l e 474   Specify the Status of an Interface
By default, an interface can both receive and send RIP update packets.
The undo rip work command is similar to undo network command in that the 
interface using either command no longer transmits an RIP route. They differ in 
that in 
undo rip work mode, routes of related interfaces are forwarded and in 
undo network mode, routes of related interfaces are not forwarded, as if an 
interface was missing. 
In addition, rip work functions similar to the combination of two commands rip 
input
 and rip output.
Disabling Host RoutesIn some special cases, a router may receive large number of host routes from the 
same network segment. These routes consume lots of network resources and are 
of little use to route addressing. You can use the 
undo host-route command to 
reject the messages of the host routes.
Perform the following configurations in RIP view.
Ta b l e 475   Disable a Host Route
OperationCommand
Enable check zero field of RIP version 1checkzero
Disable check zero field of RIP version 1undo checkzero
OperationCommand
Specify running RIP on the interfacerip work
Disable running RIP on the interfaceundo rip work
Specify receiving RIP update packets on 
the interfacerip input
Disable receiving RIP update packets on 
the interfaceundo rip input
Specify sending RIP update packets on the 
interfacerip output
Disable transmitting RIP updated packets 
on the interfaceundo rip output
OperationCommand 
						

							418CHAPTER 28: CONFIGURING RIP
By default, the router is enabled to receive the host routes.
Enabling Route 
Summarization for RIP 
Version 2Route summarization summarizes the routes of different subnets within the same 
natural network segment and sends the summary to other network segments as a 
summarized route with a natural mask. Route summarization largely reduces the 
network expenditure and the routing table size.
RIP-1 always sends routes with natural mask. RIP-2 supports sub-net mask and 
routs of unknown category. If the sub-net route needs to be broadcast, RIP-2 route 
summary function can be disabled.
Perform the following configurations in RIP view.
Ta b l e 476   Enable Route Summarization
By default, RIP-2 automatic route summarization is enabled.
Configuring RIP-2 Packet 
Authentication on the 
InterfaceAuthentication for packets is not supported by RIP Version 1. But RIP Version 2 
supports authentication.
RIP Version 2 supports authentication in two modes: simple text authentication 
and MD5 authentication. Security is not ensured in simple text authentication. 
Simple text means that the unencrypted authentication is transmitted with the 
packets, therefore simple text authentication does not apply to a situation that 
requires a high level of security. MD5 authentication has two message formats, in 
compliance of the requirements of RFC1723 (RIP Version 2 Carrying Additional 
Information) and RFC2082 (RIP Version 2 MD5 Authentication).
3Com routers support both formats.
Perform the following configurations in interface view.
Ta b l e 477   Configure Authentication for RIP Version 2
By default, RIP Version 2 packets are not authenticated on an interface. If MD5 
authentication type is not specified, the nonstandard authentication type is used 
by the router.
Disable receiving host routesundo host-route
Enable receiving host routeshost-route
OperationCommand
Enable automatic route summarizationsummary
Disable the automatic summarization 
function of RIP-2undo summary
OperationCommand
Specify a password for RIP Version 2 
simple text authenticationrip authentication-mode simple 
password
Specify a key-string for RIP Version 2 MD5 
authenticationrip authentication-mode md5 
key-string string
Set the packet format type of RIP-2 MD5 
authentication rip authentication-mode md5 type [ 
nonstandard | usual ]
Cancel authentication for RIP Version 2undo rip authentication-mode 
						

							Configure RIP 419
Configuring RIP 
Horizontal 
Segmentation on the 
InterfaceRIP is a Distance-Vector algorithm routing protocol. It uses the split-horizon 
algorithm to avoid loop routes. Split-horizon means that routes received at a 
certain interface are not sent to the same interface. If correct transmission of 
routes is more important than efficiency, then split-horizon should be disabled.
Disabling split-horizon mechanism is not effective on point-to-point connection 
links. 
Perform the following configurations in interface view.
Ta b l e 478   Configure RIP Horizontal Segmentation on the Interface
By default, the interface use split-horizon when sending RIP packets.
Configuring Route 
Import for RIPRIP allows importing the routes learned from other protocols.
Perform the following configurations in RIP view.
Ta b l e 479   Configure Route Import for RIP
By default, RIP does not import routes from other domains into the routing table.
The protocol attribute specifies the source routing domain that can be imported. 
At present RIP can import routes domain such as Connected, Static, OSPF, 
OSPF-ASE, and BGP.
See“Configure Route Import” in “Configuration of IP Routing Policy” for the 
details of importing routes.
Specifying Default 
Route Metric Value for 
RIPThe import-route command is used to import routes of other routing protocols. 
If 
import-route is not followed by the value of a routing metric, then the 
parameter value of 
default-med command is set as the metric value when 
distributing other routing protocols.
Perform the following configurations in RIP view.
Ta b l e 480   Specify a Default Route Metric Value for RIP
By default, the default route metric for RIP is 16
OperationCommand
Configure RIP horizontal segmentation on 
the interfacerip split-horizon
Prohibit the interface from using 
split-horizon when sending RIP packets.undo rip split-horizon
OperationCommand
Configure route import for RIPimport-route protocol [ cost cost ] [ 
route-policy policy-name ]
Cancel route distribution for RIPundo import-route protocol
OperationCommand
Specify default route metric value for RIPdefault-cost cost
Restore the default route metric value for 
RIP undo default-cost 
						

							420CHAPTER 28: CONFIGURING RIP
Since the route metric of route import cannot be reverted, the dynamic route 
information may be significantly distorted. Therefore, route import is done 
cautiously to prevent loss of RIP protocols performance.
Specifying Additional 
Route Metric Values for 
RIPThe additional routing metric here is to add input or output metric for routes 
obtained for RIP. The
 rip metricin will add a designated metric value while 
receiving routes on the interface, then add this route metric value in the routing 
table. The 
rip metricout does not directly change the route metric value in the 
routing table, but will add a designated metric value when sending routes on the 
interface.
Perform the following configurations in interface view.
Ta b l e 481   Specify Additional Route Metric Value for RIP
By default, the additional route metric value received for RIP is 0 but ranges from 0 
to 16. Additional route metric value being advertised for RIP is 1, ranging from 1 
to 16. 
Setting Route PreferenceEach routing protocol has its own preference that decides which routing protocol 
is used to select the best route by IP route strategy. The greater the value is, the 
lower the preference. RIP preference can be set manually.
Perform the following configurations in RIP view.
Ta b l e 482   Set Route Preference
By default, the RIP route preference is 100.
Configuring Route 
Distribution for RIPPerform the following configurations in RIP view.
Configure filtering route information received by RIP.
Ta b l e 483   Filter Routing Information Received by RIP
OperationCommand
Specify additional route metric value 
received for RIPrip metricin metric
Restore the additional route metric value 
received for RIP to its default valueundo rip metricin
Specify additional route metric value being 
advertised for RIPrip metricout metric
Restore the additional route metric value 
being advertised for RIP to its default valueundo rip metricout
OperationCommand
Set the RIP route preference preference value
Restore the default value of RIP route 
preferenceundo preference
OperationCommand
Filter routing information received from a 
specified gatewayfilter-policy gateway 
prefix-list-name import 
						

							Displaying and Debugging RIP 421
Configure filtering the routing information being advertised
Ta b l e 484   Filter the Routing Information Being Advertised by RIP
By default, RIP does not filter any route information received or being advertised.
The protocol attribute specifies the routing domain that can be filtered. At 
present, RIP can filter routes domain such as Connected, Static, OSPF, OSPF-ASE 
and BGP. See “Configure Route Filter” of “Configuration of IP Routing Policy” for 
details.
Resetting RIPThis command restores the router to the default RIP configuration.
Perform the following configuration in RIP view.
Ta b l e 485   Reset RIP
Displaying and 
Debugging RIP Ta b l e 486   Display and Debug RIP
Change or cancel filtering the routing 
information received from a specified 
gatewayundo filter-policy gateway 
prefix-list-name import
Filter the routing information receivedfilter-policy {acl-number | ip-prefix 
prefix-list-name } import
Change or cancel filtering routing 
information receivedundo filter-policy {acl-number | 
ip-prefix prefix-list-name } import
Filter routing information received from a 
specified gateway and the routing 
information received according to 
prefix-listfilter-policy ip-prefix 
prefix-list-name gateway 
prefix-list-name import
Change or cancel filtering the routing 
information received from a specified 
gateway and the routing information 
received according to prefix-listundo filter-policy ip-prifix 
prefix-list-name gateway 
prefix-list-name import
OperationCommand
Filter the routing information being 
advertised.filter-policy { acl-number | 
ip-prefix prefix-list-name } export [ 
protocol ]
Change or cancel filtering the routing 
information being advertisedundo filter-policy { acl-number | 
ip-prefix prefix-list-name } export [ 
protocol ]
Operation Command
OperationCommand
Reset RIPreset
OperationCommand
Display current RIP running status and 
global configuration information.display rip
Turn on RIP debugging informationdebugging rip { packets | receive | 
send } 
						

							422CHAPTER 28: CONFIGURING RIP
RIP - Unicast 
Configuration 
ExampleRIP is a broadcast protocol so it can only exchange routing information with 
non-broadcasting networks in unicast mode. This example shows how to 
configure RIP message unicasting.
Router A connects Router B and Router C with serial lines in non-broadcasting 
networks. Router A (192.1.1.2) only wants to send the routing updating 
information to the adjacent Router B (192.1.1.2) without sending the information 
to Router C.
Figure 145   Networking diagram of configuring the RIP unicast
To configure RIP unicast:
1Configure RIP on Router A:
[RouterA] rip
[RouterA-rip] network 192.1.1.0
2Configure Router As unicast peer to be Router B.
[RouterA-rip] peer 192.1.1.2
3Configure serial interface Serial 0
[RouterA-rip] interface serial 0
[RouterA-Serial0] ip address 192.1.1.1 255.255.255.0
Troubleshooting RIP No updating messages can be received when the physical connection works well.
This may have the following cause:
■RIP is not running on the corresponding interface (maybe the undo rip work 
command has been executed), or this interface is not included with the 
network command. Multicasting has been configured on the opposite router 
(perhaps 
rip version 2 multicast command has been executed), but not 
configured on the local router.
Router A
Router B Router CRouterA unicasts the
upgraded routing
information to RouterB s0 192.1.1.1
s0 192.1.1.2 
						

							29
CONFIGURING OSPF
This chapter covers the following topics:
■OSPF Overview
■Configuring OSPF
■Displaying and Debugging OSPF
■OSPF Configuration Example
OSPF OverviewOpen Shortest Path First (OSPF) is an autonomous, link-state-based internal 
routing protocol developed by Internet Engineering Task Force (IETF). The current 
version is version 2 (RFC1583), which features the following: 
■Applicable range — Supports networks of various sizes and hundreds of 
routers.
■Fast convergence — Sends an update message immediately after the 
topological structure of the network is changed, so the change can be 
synchronized in the autonomous system.
■No self-loop — OSPF calculates the route with the shortest path tree algorithm 
through the collected link status. This algorithm ensures that no self-loop route 
is generated.
■Area division — An AS network can be divided into areas and the routing 
information between the areas is further abstracted, reducing the bandwidth 
occupation in the network.
■Equivalent route ----support multiple equivalent routes to the same destination 
address.
■Route level --- the four levels of routes according to different priorities: 
intra-area routes, inter-area routes, external route class 1 and external route 
class 2.
■Authentication ---- support interface-based message authentication to ensure 
the security of the route computation.
■Multicast ---packets are transmitted and received with multicast address on 
multicasting link layer, greatly reducing interference to other netw ork devices.
The entire network is composed of multiple autonomous systems (AS). The link 
state of an AS is collected and transmitted to determine and propagate the route 
dynamically and then synchronize the information of the AS. Each system is 
divided into areas. If a router port is allocated to multiple areas, it is an area 
boundary router (ABR) since it is located at the boundary and connected with 
multiple areas. Routing information of another area can be learned from the ABR. 
All ABRs and the routers between them form a backbone area, tagged with  
						

							424CHAPTER 29: CONFIGURING OSPF
0.0.0.0. All areas must be continuous logically. Thus, a virtual link is introduced to 
the backbone to ensure that physically separated areas are still connected logically. 
The router between the ASs is called autonomous system boundary router (ASBR). 
Routing information, such as static routing, RIP routing, BGP routing, outside the 
OSPF AS can be learned from the ASBR.
Computation of the OSPF protocol is summarized as follows:
1Every router supporting OSPF maintains a link state database (LSDB) for describing 
the topology of the entire AS. A router generates the Link State Advertisement 
(LSA) according to the network topology around it and sends the LSA to other 
routers on the network by the transmission of protocol packets. Thus, every router 
receives the LSA from other routers. All LSAs together forms the LSDB. 
2The LSA describes the network topology around a router, so the LSDB describes 
the topology of the whole network. A router can easily convert the LSDB into a 
weighted directed graph, which shows the real topology of the whole network. 
Obviously, each router in the autonomous system receives the same topology 
diagram of the network.
3Each router calculates with the SPF algorithm a shortest path tree with itself as the 
root. This tree gives the routes to all autonomous systems. External routing 
information is the leaf sub-node. The external route is flagged by the router by 
broadcasting it to record additional information for the AS. Obviously, each router 
gets a different routing table.
In addition, multiple adjacent relationship lists must be created so that each router 
on the broadcast network and NBMA network can broadcast the local status 
information (such as available interface information and reachable peer 
information) to the whole system. Consequently, the route change of any router 
may be transmitted many times, which is both unnecessary and wastes bandwidth 
resources. To solve this problem, OSPF protocol selects a designated router (DR). 
All routers send information to the DR, which broadcasts the network link status. 
Two non-DR routers (DR Other) do not create neighboring relations with each 
other and do not exchange any routing information. Then the number of 
neighboring relations between the routers on the multi-address network is greatly 
The OSPF protocol supports IP subnet and the marking and receiving of external 
routing information. It supports interface-based message authentication to insure 
the security of route calculation. Messages are transmitted and received in IP 
multicast mode.
Configuring OSPFIn all configuration tasks, the OSPF-specified interface and area number must be 
defined first to configure other function features. The configuration of 
interface-related function features is not restricted by whether OSPF has been 
enabled. The original interface parameters become invalid after OSPF is 
terminated.
OSPF configuration includes:
■Specify Router ID
■Enabling OSPF
■Associating an Area-id with the Specified Interface
■Configuring the Network Type of the OSPF Interface  
						

							Configuring OSPF425
■Configuring Sending Packet Cost
■Configuring a Peer for the NBMA Interface
■Specifying the Router Priority
■Specifying the Hello Interval
■Specifying the Dead Interval
■Specifying the Retransmitting Interval
■Specifying the Transmit-delay
■Configuring a Stubby Area and a Totally Stubby Area
■Configuring an NSSA Area
■Configuring Route Summarization within the OSPF Domain.
■Creating and Configuring a Virtual Link
■Configuring Authentication
■Configuring Route Import for OSPF
■Configuring Parameters when Importing External Routes
■Setting Route Preference
Specify Router IDRouter ID is a 32-bit integral with symbol, the exclusive ID of a router in the AS. If 
all interfaces of the router have not been configured with IP addresses, the router 
ID must be configured in OSPF view, otherwise OSPF will not run. 
The modified router ID takes effect after OSPF is restarted.
You must configure the router ID, which must be the same as the IP address of a 
specific interface of this router.
Perform the following configurations in system view.
Ta b l e 487   Specify Router ID
Please note when modifying the router ID, the system will display the following 
message:
OSPF: router id has changed. If you want to use new router id, reboot 
the router. 
The configuration needs to be saved after the router ID is modified (execute the 
save command in system view). After restarting the router, the new router ID will 
take effect.
Enabling OSPFPerform the following configurations in system view.
Ta b l e 488   Enable OSPF
OperationCommand
Specify the router IDrouter id router-id
Delete the router IDundo router id 
OperationCommand 
						

							426CHAPTER 29: CONFIGURING OSPF
By default, OSPF is disabled.
Associating an Area-id 
with the Specified 
InterfaceThe OSPF protocol divides the autonomous system into areas. An area is the 
logical group of the router. Some routers belong to different areas (called area 
boundary router ABR), while a network segment can only be in one area. In other 
words, each interface running the OSPF protocol must be put in a specific area. 
The area is flagged with an area ID. The ABR transmits routing information 
between areas.
In addition, in the same area, all routers must agree unanimously to the parameter 
configurations of this area. So, in the configuration of routers in the same area, 
most configuration data must be considered on the basis of this area. Incorrect 
configurations make it impossible for adjacent routers to transfer information to 
each other, or can even lead to the blocking or self-loop of routing information.
Perform the following configurations in interface view.
Ta b l e 489   Associate an Area-id with the Specified Interface which runs OSPF
No area-id is associated with the specified interface by default after OSPF is 
enabled.
After OSPF is enabled, you must specify an area-id associated with the specified 
interface. OSPF only works on the specified interface.
Configuring the 
Network Type of the 
OSPF Interface The OSPF protocol calculates the route on the basis of the topological structure of 
the neighboring network of this router. Each router describes the topology of its 
neighboring network and transmits this information to all other routers.
OSPF divides the network into 4 types according to the link layer protocols:
■When the link layer is Ethernet, OSPF regards the network type as broadcast by 
default.
■When the link layer protocol is frame relay, HDLC and X.25, OSPF regards the 
network type as NBMA by default.
■No link layer protocol is considered as point-to-multipoint type by default. It is 
usually manually modified from NBMA if the NBMA network is not wholly 
interconnected. 
■When the link layer protocol is PPP, LAPB, OSPF regards the network type as 
point-to-point by default.
NBMA is a Non Broadcast Multi Access network. The typical network is X.25 and 
frame relay. Configure the poll-interval to specify the period for sending a polling 
Enable OSPF and enter into the OSPF viewospf enable
Turn off OSPFundo ospf enable
OperationCommand
Specify an area-id associated with the 
specified interface which runs OSPFospf enable area area-id
Delete an area-id associated with the 
specified interfaceundo ospf enable area area-id