Lab 28: DLSw Reachability, Border Peers, Demand Peers, and Resilient Peers-Part I

Integrated Routing and Bridging

Integrated routing and bridging (IRB) allows you to bridge local traffic within several segments while having hosts on the bridged segments reach the hosts or routers on routed networks. It essentially can allow a routed domain to reach a bridge domain.

Using the IRB feature, you can route a given protocol between routed interfaces and bridge groups within a single router. Specifically, local or unroutable traffic will be bridged among the bridged interfaces in the same bridge group , while routable traffic will be routed to other routed interfaces or bridge groups.

Integrated routing and bridging uses the concept of a Bridge-Group Virtual Interface (BVI) to enable these interfaces to exchange packets for a given protocol. A BVI is a virtual interface within the router that acts like a normal routed interface. A BVI does not support bridging, but it actually represents the corresponding bridge group to routed interfaces within the router. The interface number is the link between the BVI and the bridge group. All Layer 3 information, such as IP address, or filters are applied to the BVI, not the actual physical interface.

IRB Considerations

Before enabling IRB, you should be aware of the following:

• The default route/bridge behavior on a router is to route all packets first and then bridge them. This is precisely why configuring transparent bridging does not impact the routed domain. However, when the IRB is enabled, the behavior changes to bridge all packets first. The bridge group must be enabled to route with command bridge bridge_number route ip if routing IP also is enabled .

• Packets of nonroutable protocols, such as local-area transport (LAT) or SNA, always are bridged. You cannot disable bridging for the nonroutable protocols.

• Bridging attributes cannot be configured on a BVI interface.

• IRB supercedes concurrent routing and bridging (CRB), which no longer should be used.

Figure 13-5 illustrates a common IRB environment. In this figure, note that there are no Layer 3 addresses on the Ethernet interfaces that are in the bridged domain. These interfaces are made part of the BVI by being members of the same bridge group. The BVI number ”in this case, 10 ”must be the bridge number. This number is how the two domains are linked. The BVI interface is where all layer information goes for the protocol that you want to bridge and route. In this figure, you simply have an IP address. By adding an IP address to the BVI interface, it instructs the router to start to bridge IP on all interfaces that are a member of bridge group 10.

Figure 13-5. Integrated Routing and Bridging

Configuring IRB

Configuring IRB is a three-step process. The steps are as follows :

Step 1. Configure transparent bridging on the interfaces that you want to bridge and route. Use the process defined earlier. Recall that this consists of creating a bridge group and assigning interfaces to that group.

Step 2. Configure IRB and the BVI. Assign the same bridge number to the BVI interface. For example if you use bridge 2, the BVI will be interface bvi 2. To enable IRB, use the following syntax from the router global configuration prompt:

 Router(config)#  bridge irb  Router(config)#  interface bvi   bridge-group_number  

Step 3. Configure protocol-specific routing and bridging for the bridge group. This step is extremely important. As soon as you enable IRB with the bridge irb command, all interfaces in that bridge group, or all interfaces that have transparent bridging enabled on them, will now bridge all protocols first. This can and will have devastating effects where routing is needed for another protocol on those links. Because of this, all Layer 3 protocols must be told whether to route, bridge, or do both. To accomplish this, you must complete two steps:

1. Assign all Layer 3 addresses of the protocol that you want to bridge and route to the BVI interface. No Layer 3 addresses should reside on the physical interface for the protocol that you want to bridge and route.

2. Enable or disable routing and bridging on a per-protocol basis.

To enable bridging or routing per protocol, use the following syntax from the global configuration prompt:

 Router(config)#  bridge   bridge_number  [  route   bridge  ] [  ip   ipx   appletalk     decnet   ]  

To disable bridging or routing per protocol, use the following syntax from the global configuration prompt:

 Router(config)#  no bridge   bridge_number  [  route   bridge  ] [  ip   ipx   appletalk     decnet   ]  

To view whether the router is routing, bridging, or both for any given protocol, use the command show interface irb, as demonstrated in Example 13-5.

Example 13-5 show interface irb Command Output

 irb_router#  show int irb  Ethernet2  Routed protocols on Ethernet2:   ip         ipx   Bridged protocols on Ethernet2:   appletalk  clns       decnet     vines   apollo     xns  Software MAC address filter on Ethernet2   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x86:  0 00e0.1e58.e798         0 RCV Interface MAC address   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree 

Practical Example: Configuring IRB

Figure 13-6 shows a network where you want to bridge and route IP on interfaces e2 and e3. To configure IRB on this router, you begin by defining a bridge group and placing the interfaces that you want to bridge and route into that bridge group. In this example, we have made bridge group 5 and placed interfaces e2 and e3 into that bridge group.

Figure 13-6. Integrated Routing and Bridging Example

If you examine IRB, at this point, it looks much like a normal router. Example 13-6 lists the output of the show interface irb command on the irb_router. Notice that the router is either bridging or routing IP, but not both.

Example 13-6 show interface irb Command Output

 irb_router#  show int irb  Ethernet2  Routed protocols on Ethernet2:   ip         ipx  Bridged protocols on Ethernet2:   appletalk  clns       decnet     vines   apollo     xns  Software MAC address filter on Ethernet2   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x86:  0 00e0.1e58.e798         0 RCV Interface MAC address   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree Ethernet3  Routed protocols on Ethernet3:   ip         ipx  Bridged protocols on Ethernet3:   appletalk  clns       decnet     vines   apollo     xns  Software MAC address filter on Ethernet3   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x85:  0 00e0.1e58.e79b         0 RCV Interface MAC address   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree Ethernet4  Routed protocols on Ethernet4:   ip         ipx  irb_router# 

The second step in the configuration involves enabling IRB and creating the BVI. Because you are using bridge 5, the BVI interface will be 5, too. When the BVI is created, the router generates the statements needed to route whatever Layer 3 protocol it discovers on a transparently bridged interface. Example 13-7 shows this happening when IRB was enabled in the model. In this model, IPX and IP are enabled on a transparently bridged interface, so both commands were generated.

Example 13-7 Enabling IRB

 irb_router(config)#  bridge irb  IRB: generating 'bridge 5 route ip' configuration command IRB: generating 'bridge 5 route novell' configuration command irb_router(config)#  int bvi 5  04:17:56: %LINEPROTO-5-UPDOWN: Line protocol on Interface BVI5, changed state to  up irb_router(config-if)# 

Now you want to remove the Layer 3 address of the protocol that we want to route and bridge from the physical interface. All other Layer 3 addresses, the ones that you do not want to bridge and route on, should remain on the physical interface. Assign the Layer 3 address of the protocol to bridge and route to the BVI interface. Figure 13-7 illustrates the changes that you need to make to the network.

Figure 13-7. Addressing and Creating the BVI

After the IP address changes have been made, the model is almost complete. At this point, IP is being bridged and routed, and this can be verified with the show int irb command, as in Example 13-8.

Example 13-8 show interface irb Command Output

 irb_router#  show int irb  BVI5  Routed protocols on BVI5:   ip  Ethernet2  Routed protocols on Ethernet2:   ip         ipx   Bridged protocols on Ethernet2:   appletalk  clns       decnet     ip   vines      apollo     ipx        xns  Software MAC address filter on Ethernet2   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x86:  0 00e0.1e58.e798         0 RCV Interface MAC address   0x86:  1 00e0.1e58.e798         0 RCV Bridge-group Virtual Interface   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree Ethernet3  Routed protocols on Ethernet3:   ip         ipx   Bridged protocols on Ethernet3:   appletalk  clns       decnet     ip   vines      apollo     ipx        xns  Software MAC address filter on Ethernet3   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x85:  0 00e0.1e58.e79b         0 RCV Interface MAC address   0x86:  0 00e0.1e58.e798         0 RCV Bridge-group Virtual Interface   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree Ethernet4  Routed protocols on Ethernet4:   ip         ipx  irb_router# 

By observing the output of Example 13-8, you can see that the BVI appears as a normal interface, much like interface E4. Notice that only IP is running on the BVI because that is the only protocol that you want to bridge and route.

Notice in this model that you have a downstream IPX router. All IPX traffic first is bridged, and this causes problems with the downstream IPX router. To remedy this, the second part of Step 3 calls for you to disable IPX bridging with the command no bridge 5 bridge ipx. After keying in this command, you can view the IRB interface and note the changes. IPX no longer is bridged and routed on the interfaces E2 and E3; it is only routed. Example 13-9 lists the output of the show irb command, illustrating that IPX no longer is bridged and routed.

Example 13-9 show interface irb Command Output

 irb_router#  show int irb  BVI5  Routed protocols on BVI5:   ip Ethernet2  Routed protocols on Ethernet2:   ip         ipx   Bridged protocols on Ethernet2:   appletalk  clns       decnet     ip   vines      apollo     xns  Software MAC address filter on Ethernet2   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x86:  0 00e0.1e58.e798         0 RCV Interface MAC address   0x86:  1 00e0.1e58.e798         0 RCV Bridge-group Virtual Interface   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree Ethernet3  Routed protocols on Ethernet3:   ip         ipx   Bridged protocols on Ethernet3:   appletalk  clns       decnet     ip   vines      apollo     xns  Software MAC address filter on Ethernet3   Hash Len    Address      Matches  Act      Type   0x00:  0 ffff.ffff.ffff         0 RCV Physical broadcast   0x2A:  0 0900.2b01.0001         0 RCV DEC spanning tree   0x85:  0 00e0.1e58.e79b         0 RCV Interface MAC address   0x86:  0 00e0.1e58.e798         0 RCV Bridge-group Virtual Interface   0xC0:  0 0100.0ccc.cccc         0 RCV CDP   0xC2:  0 0180.c200.0000         0 RCV IEEE spanning tree   0xC2:  1 0180.c200.0000         0 RCV IBM spanning tree Ethernet4  Routed protocols on Ethernet4:   ip         ipx  irb_router# 

Example 13-10 lists the configuration that we created for this model.

Example 13-10 IRB Configuration for This Model

  hostname irb_router   !   ip subnet-zero   ipx routing 00e0.1e58.e792   !   bridge irb   !   <<<text omitted>>>   !   interface Ethernet2   no ip address   no ip directed-broadcast   media-type 10BaseT   ipx network 10   bridge-group 5   !   interface Ethernet3   no ip address   no ip directed-broadcast   media-type 10BaseT   ipx network 20   bridge-group 5   !   interface Ethernet4   ip address 172.16.3.1 255.255.255.0   no ip directed-broadcast   media-type 10BaseT   ipx network 30   !   <<<text omitted>>>   !   interface BVI5   ip address 172.16.1.1 255.255.255.0   no ip directed-broadcast   !   router eigrp 2001   network 172.16.0.0   !   ip classless   !   bridge 5 protocol ieee   bridge 5 route ip   bridge 5 route ipx   no bridge 5 bridge ipx   !