Section 3: ISDN (8 Points)

• Ensure that VLAN3 and R4 Lo0 are accessible from R1 and beyond should the Frame Relay network fail either physically or logically. If VLAN3 and R4 Lo0 networks are restored while the ISDN line is active, ensure that traffic is routed over the Frame Relay network to these destinations immediately.

As no static routes are permitted and backing up the Frame Relay interface will not help as this only works if the Frame Relay interface is physically down, the only option will be to use the dialer-watch feature. Both networks must be down before the router dials out so VLAN3 and R4 Lo0 should be added to a dialer watch-list and corresponding dialer watch-group number under the BRI interface on R1. R1 is used to dial out as the question states that the two networks should be accessible from R1. You are also later advised that only R1 should dial into R4.

You should notice that when you fail the Frame Relay network to test this that after the ISDN is activated and the Frame Relay network is then restored that the routing table on R1 shows identical hop counts for all remote networks via R4 over both the Frame Relay and ISDN line as shown in Example 1-23.

This condition can keep the ISDN line from ever deactivating as the ISDN network can now be used as a valid means to transport data to the RIP advertised remote networks, you should also notice that the question requires that the Frame Relay routes should be used "immediately" when restored and, at this point, routers R1 and R4 can choose between Frame Relay and ISDN.

RIP obviously does not take into account the bandwidth of available routes. You, therefore, need to make the ISDN routes less desirable and add additional hop count to RIP using an offset-list on R4 and R1 out over the ISDN line (inbound over both routers will also be acceptable). This ensures when the Frame Relay is restored and for the period where both Frame Relay and ISDN lines are active and receiving RIP routes that the hop count is more favorable over Frame Relay because of the additional hop count incurred over ISDN after the offset-list is applied.

The ISDN line can not be used to route traffic while a higher-speed Frame Relay connection is available as shown in Example 1-25. Dialer-watch does not require interesting traffic to trigger the dial so the dialer-list should be an implicit deny of any IP traffic; otherwise, any traffic will potentially keep the line up after initiated. It is better practice and shows a better understanding of the dialer-watch process to, therefore, have the following dialer-list on R1; dialer-list 10 protocol ip deny. You will find with this strict policing of the interesting traffic, your ISDN line will stay down when the networks are restored over the Frame Relay. If you have configured this question correctly as in Example 1-24, you have scored 5 points; if you have used a dialer-list that denies RIP and the line stays down, you have only scored 3 points. Test your scenario thoroughly if you have first denied RIP then allowed all other IP traffic and also not applied the offset-list; you could find that with two routes in the routing table with identical metrics that traffic, such as BGP, will toggle between the two routes and keep the line up constantly. In addition, other IP traffic could be classed as interesting and keep the line up.

Example 1-23. R1 Routing Table Pre Offset-List with the ISDN Line Active After the Frame Relay Network Has Been Restored

 R1#sh ip route R    172.16.0.0/16 [120/1] via 10.90.90.1, 00:00:22, Serial0/0      10.0.0.0/8 is variably subnetted, 16 subnets, 5 masks R       10.8.8.8/32 [120/4] via 10.10.10.2, 00:00:14, BRI0/0                     [120/4] via 10.100.100.2, 00:00:14, Serial0/1.101 C       10.10.10.2/32 is directly connected, BRI0/0 C       10.100.100.0/28 is directly connected, Serial0/1.101 R       10.99.99.0/29 [120/2] via 10.10.10.2, 00:00:14, BRI0/0                       [120/2] via 10.100.100.2, 00:00:14, Serial0/1.101 R       10.60.60.0/29 [120/2] via 10.10.10.2, 00:00:14, BRI0/0                       [120/2] via 10.100.100.2, 00:00:14, Serial0/1.101 R       10.50.50.0/29 [120/4] via 10.10.10.2, 00:00:14, BRI0/0                       [120/4] via 10.100.100.2, 00:00:14, Serial0/1.101 R       10.40.40.0/28 [120/1] via 10.10.10.2, 00:00:16, BRI0/0                       [120/1] via 10.100.100.3, 00:00:16, Serial0/1.101 R       10.7.7.0/28 [120/4] via 10.10.10.2, 00:00:16, BRI0/0                     [120/4] via 10.100.100.2, 00:00:16, Serial0/1.101 R       10.6.6.0/29 [120/2] via 10.10.10.2, 00:00:16, BRI0/0                     [120/2] via 10.100.100.2, 00:00:16, Serial0/1.101 R       10.4.4.0/29 [120/1] via 10.10.10.2, 00:00:16, BRI0/0                     [120/1] via 10.100.100.3, 00:00:16, Serial0/1.101 C       10.80.80.0/24 is directly connected, FastEthernet0/0 C       10.90.90.0/28 is directly connected, Serial0/0 C       10.1.1.0/28 is directly connected, Loopback0 C       10.10.10.0/28 is directly connected, BRI0/0 C       10.90.90.1/32 is directly connected, Serial0/0 R       10.5.5.4/30 [120/4] via 10.10.10.2, 00:00:16, BRI0/0                     [120/4] via 10.100.100.2, 00:00:16, Serial0/1.101 R1#sh isdn history --------------------------------------------------------------------------------                                 ISDN CALL HISTORY -------------------------------------------------------------------------------- Call History contains all active calls, and a maximum of 100 inactive calls. Inactive call data will be retained for a maximum of 15 minutes. -------------------------------------------------------------------------------- Call    Calling      Called       Remote  Seconds Seconds Seconds Charges Type    Number       Number       Name    Used    Left    Idle    Units/Currency -------------------------------------------------------------------------------- Out                    2222                    82     37       82      0 -------------------------------------------------------------------------------- 

NOTE

The routing table output is taken after a Frame Relay failure is restored and the ISDN line is still active. The shading shows you the two available routes with the identical hop count on R1 before the offset-list is applied.

Example 1-24. Increasing the Hop Count Out of R4 and R1 ISDN Configuration

 R1 router rip offset-list 0 out 2 BRI0/0 ________________________________________________________________ R4 router rip  offset-list 0 out 2 Dialer0 

NOTE

offset-list 0 will apply the chosen additional hop count (2) to all networks being advertised from R4 and R1 out of their interfaces BRI0/0. A similar configuration could be placed on each BRI0/0 but inbound.

Example 1-25. R1 Routing Table Post Offset-List with the ISDN Line Active After the Frame Relay Network Has Been Restored

 R1#sh ip route R    172.16.0.0/16 [120/1] via 10.90.90.1, 00:00:09, Serial0/0      10.0.0.0/8 is variably subnetted, 16 subnets, 5 masks R       10.8.8.8/32 [120/4] via 10.100.100.2, 00:00:28, Serial0/1.101 C       10.10.10.2/32 is directly connected, BRI0/0 C       10.100.100.0/28 is directly connected, Serial0/1.101 R       10.99.99.0/29 [120/2] via 10.100.100.2, 00:00:28, Serial0/1.101 R       10.60.60.0/29 [120/2] via 10.100.100.2, 00:00:28, Serial0/1.101 R       10.50.50.0/29 [120/4] via 10.100.100.2, 00:00:28, Serial0/1.101 R       10.40.40.0/28 [120/1] via 10.100.100.3, 00:00:28, Serial0/1.101 R       10.7.7.0/28 [120/4] via 10.100.100.2, 00:00:28, Serial0/1.101 R       10.6.6.0/29 [120/2] via 10.100.100.2, 00:00:28, Serial0/1.101 R       10.4.4.0/29 [120/1] via 10.100.100.3, 00:00:28, Serial0/1.101 C       10.80.80.0/24 is directly connected, FastEthernet0/0 C       10.90.90.0/28 is directly connected, Serial0/0 C       10.1.1.0/28 is directly connected, Loopback0 C       10.10.10.0/28 is directly connected, BRI0/0 C       10.90.90.1/32 is directly connected, Serial0/0 R       10.5.5.4/30 [120/4] via 10.100.100.2, 00:00:00, Serial0/1.101 R1#sh isdn hist --------------------------------------------------------------------------------                                 ISDN CALL HISTORY -------------------------------------------------------------------------------- Call History contains all active calls, and a maximum of 100 inactive calls. Inactive call data will be retained for a maximum of 15 minutes. -------------------------------------------------------------------------------- Call    Calling      Called       Remote  Seconds Seconds Seconds Charges Type    Number       Number       Name    Used    Left    Idle    Units/Currency -------------------------------------------------------------------------------- Out                    2222                    92     27       92      0 -------------------------------------------------------------------------------- 

NOTE

The routing table output is taken after a Frame Relay failure is restored and the ISDN line is still active. This shows that the ISDN routes are no longer entered into the routing table on R1 because of the increased hop count over this environment. The routing table on R4 will act in exactly the same manner.

• Configure R1 so that if half of the ISDN traffic to R4 is of an unacceptable quality, the line is automatically disconnected.

Configure ppp quality 50 under both R1 and R4 BRI0/0 interfaces, the figure (percentage) is for both incoming and outgoing directions on the interface, PPP will drop the line if the quality falls below 50 percent and initiate a timer before re-establishing the link. If you have configured this correctly, you have scored 1 point.

• Allow only R1 to dial into R4. Do not use any PPP feature in your solution.

The question is not seeking configuration of CHAP on both routers as any router configured with the correct CHAP password could emulate R1 and gain access to R4. It is, therefore, required to configure R4 with isdn caller 1111 if using legacy DDR or dialer-caller 1111 if using dialer profiles to ensure that only R1, which is connected to the ISDN number 1111, can actually gain access by having R4 check the CLI before answering. You may have automatically assumed this must require CHAP but there is not sufficient detail in the question to suggest that CHAP or PAP is required. These are both also PPP features so it is disallowed anyway. If you have configured this correctly, you have scored 2 points.

NOTE

Your ISDN line or simulator must support CLI to test this feature.

• Do not allow the ISDN to flap if the Frame Relay network goes up and down; only allow the line to be dropped if the Frame Relay connectivity is deemed to be reliable for 90 seconds.

By default, the ISDN line will be dropped when dialer-watch again has visibility if the networks listed in the dialer watch-list. To ensure the line remains active for 90 seconds the command dialer watch-disable 90 should be added to the BRI0/0 interface of R1. If you have configured this correctly, you have scored 1 point.

Example 1-26 and Example 1-27 show the full final ISDN and relevant RIP configuration required for the ISDN backup on R1 and R4, using a mix of legacy and dialer profile commands.

Example 1-26. R1 Final ISDN and Relevant RIP Configuration

 interface BRI0/0  ip address 10.10.10.1 255.255.255.240  encapsulation ppp  dialer watch-disable 90  dialer string 2222  dialer watch-group 5  dialer-group 10  isdn switch-type basic-net3  no peer neighbor-route  ppp quality 50 ! router rip  version 2  passive-interface default  no passive-interface BRI0/0  offset-list 0 out 2 BRI0/0  network 10.0.0.0 ! dialer watch-list 5 ip 10.4.4.0 255.255.255.240 dialer watch-list 5 ip 10.40.40.0 255.255.255.240 dialer-list 10 protocol ip deny 

Example 1-27. R4 Final ISDN and Relevant RIP Configuration

 interface BRI0/0 encapsulation ppp isdn switch-type basic-net3 dialer pool-member 1 ! interface Dialer0  ip address 10.10.10.2 255.255.255.240  encapsulation ppp  dialer pool 1  dialer-group 10  ppp quality 50  dialer-caller 1111 ! router rip  version 2  passive-interface default  no passive-interface Serial0/0.1  no passive-interface Dialer0  offset-list 0 out 2 D0  dialer-list 10 protocol ip permit