EIGRP Comprehensive Lab Notes

650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1421972700374 "border=" 0 "alt=" 1421972700374 "src=" http://s3.51cto.com/wyfs02/M00/5B/21/ Wkiol1uajzjt23osaafe3x7odem625.jpg "longdesc="./1421972700374.png "height=" 403 "/>
The actual topology should look like this:
650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1421896034445 "border=" 0 "alt=" 1421896034445 "src=" http://s3.51cto.com/wyfs02/M01/5B/21/ Wkiol1uajznat05zaadpuwlxazm239.jpg "longdesc="./1421896034445.png "height=" 247 "/>
GNS3 more easily understood topology is this:
650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1421973585243 "border=" 0 "alt=" 1421973585243 "src=" http://s3.51cto.com/wyfs02/M02/5B/21/ Wkiol1uajzmt9gn6aadneboe3bw626.jpg "longdesc="./1421973585243.png "height=" 311 "/>

Frame-relay

1. : With Frame-relay SWITCH, the Frame-relay PVC between R2.R3.R4 is Full-mesh. But use only the PVC shown in the figure.

   As can be seen from the figure, only 4 DLCI is allowed, so it needs to be configured as Hub-spoke mode. Where R4 is the hub

2. R2,R3,R4 's physical interface connects Frame-relay SWITCH,

3. Between R4 and R5. Please use a PPP link. Eliminate redundant 32-bit host routing

   PPP Link

   PPP pushes its own host routing, is to ensure that there is no address on the interface to borrow the back of the network segment address, push to the other side, to ensure that the two ends of the same network segment can communicate normally.
   Command line: Turn off host route push no peer Neighbor-route

4. R1,r2. R3 's E0 mouth. All received a hub.

5. Configure the IP address as shown in the figure.

6. Make sure that each link can ping each other.

7. . R1-r5 each have a LO0, the address is: 10.10.x.x/24 (X is your router number),

Part 1-rip

1. R1, R2, R3 runs RIP VERSION 2 and announces the direct-connected Ethernet segment of LO0 and routers to rip

Ethernet Network Segment 1.1.123.0/27

2 in the routing table of three routers, you can see each other's LO0,

3 on R1, use the smallest command line. Add eight interface addresses. The address is: 100.1.1.0/24 ... 100.1..8.0/24, these eight addresses appear as external routes in the R1 RIP database.

Using the secondary address, under the loopback 0 interface:
IP address 100.1.1.1 255.255.255.0 secondary
IP address 100.1.2.1 255.255.255.0 secondary
IP address 100.1.3.1 255.255.255.0 secondary
IP address 100.1.4.1 255.255.255.0 secondary
IP address 100.1.5.1 255.255.255.0 secondary
IP address 100.1.6.1 255.255.255.0 secondary
IP address 100.1.7.1 255.255.255.0 secondary
IP address 100.1.8.1 255.255.255.0 secondary
appears as an external route: using a redistribution direct connection

4 only in R1 do the configuration, and use the smallest command line, so that R3 can see the seven of these eight network segments, see 100.1.8.0/24 This route, and the odd route hop is 5, even route hop is 7, can not be used Offset-list

You can use Route-map to solve

Access-list 1 Permit 100.1.8.0access-list 2 permit 100.1.0.0 0.0.6.0route-map HOP permit match IP address 2 set metric 7route-map Hop Permit Match IP address 1 set metric 16route-map HOP Permit 30redistribute connected metric 5 Route-map HOP

5 can not make any configuration in R2, make R2. You cannot see these eight routes, but you can see the LO0 of R1,R3. Resolve this issue without version 1, and then resolve the issue.

Do not configure on R2, use passive interface to make R1 not send updates to R2, and then use NEI command Unicast and R3 to pass routes. However, it is important to note that horizontal segmentation is turned off on the R3 f0/0 interface due to the problem of horizontal partitioning.
Requires No 8 routes on the R2, the associated entries are filtered out on the R3 using the distribution list.

R1 (config-router) #passive-interface defaultR1 (config-router) #nei 1.1.123.3r3 (config) #acc deny 100.1.0.0 0.0.7.0R3 (config) #acc per anyR3 (config-router) #distribute-list RIP

Part 2–EIGRP 100

1. R2,R3,R4 's Frame-relay link runs eigrp, R4 's LO0 (10.10.4.4) is also declared to EIGRP 100
   On the R4. SH IP EI NEI. You can see two adjacency relationships.

2. The address of the E0 port of R4 is: 150.1.y.1/24. Make R4 with BB2 to establish EIGRP adjacency, R4 will receive a lot of routes from BB2. Include: A.B.C class. With the smallest prefix command line. Allow only Class B routing to enter R4
   Actual matching entries

   Prefix-list

   Used to filter the routing prefix
   IP prefix-list B Permit 128.0.0.0/2 le 32
   Ge-len=m
   Le-len=n
   Number of route entries that are actually matched: m+1)

3. The two-way redistribution of rip and EIGRP is done on R2,R3. Any route to the 100.0.0.0 segment is not visible on the R4. This solution cannot be used with Route-map. Nor can you use interface-based filtering, which cannot be done by means of metric.

   Route redistribution

   To pour the data of a routing protocol into the database of the B routing protocol
   Principle:

• In the redistribution, you need to write the metric.

• Cannot do interface-based filtering, so we can use distribution lists for protocol-based filtering.

• Only a protocol route that successfully submits the routing table will be re-distributed into the database of B

• A routing entry for a direct connection that is included in the B protocol but not included by the A protocol

RIP protocol: Redistribute EIGRP metric 1 EIGRP protocol: Redistribute RIP metric 100000 255 1, distribute-list out RIP

Distribute-list only out to be able to do protocol-based filtering, there is no way to do protocol-based filtering. The source route is not known because it is in.

1. Use only one instruction to make EIGRP hello=5 hold between R2,r3,r4 time=15

   Under normal circumstances can use IP hello-interval and IP hold-time under the interface to modify, when the bandwidth is less than 1544 will use 60/180 of the time, so the subject only need to increase the bandwidth so that it is greater than 1544

2. R2. R4. Requires extremely high security. Please implement.

   History Size: Modify the entries saved by history

• Open Certification

Key Chain a key 1 key-string ccie in S1/1 IP authentication mode EI MD5 IP authentication key-chain EIGRP

1. On the R2. Add LO1. The IP address is: 161.1.1.1/24, declared into EIGRP, at this time the R4 should only see 161 segments of three routes: 161.1.0.0/16, 161.1.0.0/18, 161.1.0.0/20 three routes. Only 161.1.1.0/24 routes can be seen on the R1. Make the appropriate configuration on the R2 only. Cannot use Route-map, interface-based filtering can be used

• Can see/16/18/20 can do summary, EIGRP Summary can be summarized multiple times
  The second step is relatively simple, only using Distribute-list filter 161.1.0.0 entries can be

Access-list deny 161.1.0.0access-list permit anyrip: Distribute-list out fastethernet0/0

1. Declares R3 's LO0 directly to EIGRP. So that R4 can see two equivalent routes, and

   declares before EIGRP, Since 10.10.3.3 is being re-distributed from the RIP database, load balancing between R2 and R3. After the declaration, 10.10.3.3 into the EIGRP database, directly through EIGRP notification to R4, management distance 90, then R4 choose R3 do a fright, no longer load, So we need to modify the management distance and metric. The simplest way to modify metric is to modify the delay of the R3 Lo0 interface.

   Modify Distance:distance 1.1.234.3 0.0.0.0 3
   1.1.234.3 Route Source Router
   3 matches the ACL used by the route entry

   /blockquote>

2. Add a LO5 on R3 with the address: 5.5.5.5/32. Also declared in RIP and EIGRP. Configure on R3 so that R4 can see this route, and the metric value is: 100000000, The next hop points to R3. When the R3-R4 link is broken. R4 can reach this route through R2. And no need for re-convergence,      
   The subject can be offset-list adjusted metric.

   offset-list 5 out 97703168 S1/1
   DV Protocol to develop a good habit, when doing the offset must be with the interface

li>

Adds a LO6 to the R3 with the address: 6.6.6.6/32. Also declared in RIP and eigrp. So that R4 can see this route, the next hop points to R2. When the R2-R4 link is broken. R4 can reach this route through R3. And do not need to be re-converged, use distance 1XX to complete this task, if you modify metric, you cannot use offset-list, you can not affect other routes

The topic can not be directly adjusted to manage the distance, distance can only be called once. But the problem is the same as the previous one, so the only thing you need to do is modify the ACL.

1. For the above question, if you need to install two non-equivalent routes how to do, if the request packet only use metric the smallest route, how to do? No need to consider ROUTE-CACHE,CEF and so on.

• Using variance 2 to enable non-equivalent load balancing

• Enable minimum traffic distribution
  traffic-share min across-interfaces

One R4 to BB2 only a default route. You can not inject routes to other routers, do not use any kind of filtering, please use the appropriate instructions to implement

Release of EIGRP default route:

• Aggregated to full 0

• Redistribution static

• Static declaration

• IP default-network
  This requirement cannot be injected into other routers, so the best way to do this is to summarize all 0 under the f0/0 interface.

Part 3-EIGRP 99

1. R4. The PPP link between R5 runs EIGRP 99. Use unicast to establish adjacency relationships. And put R5 's LO0 into the EIGRP process.

Normal establishment, no pits

1. Add a LO1 on the R5. The address is: 7.7.7.7/32. Make it visible in R4 as an external route:
   7.7.7.7/32 (90/xxxxx). R4 to inject only one default route to R5. Can not use Interface-summ, also can not use distribute, can not use Route-map.

• For R5 only static, a stub router can be made between R4 and R5, and only accept static

• When it comes to redistribution, it is more mature to write a route-map to match the routing entries that need to be re-distributed.

• External routes can only modify the global distance

• Static redistribution

On the R4. Two-way redistribution of EIGRP 100 and EIGRP 99 is shown on the R1. You only see a 7.7.7.7/32 route with the next hop pointing to R2. If the R2 is unreachable. Just R1 can reach this route via the R3. You should be able to ping this route on the R1. Note that R2 and R3 can pass routes to each other, and this scenario may need to consider preventing the routing ring, but not with offset-list, and R1 cannot modify distance.

How to avoid suboptimal paths in scenarios where bidirectional double-point redistribution is avoided

The essence of the suboptimal path is that after the B protocol is poured into the a protocol, the B zone is poured back into the a protocol.
Use route map to set tag to filter routes
650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1422006929911 "border=" 0 "alt=" 1422006929911 "src=" http://s3.51cto.com/wyfs02/M00/5B/21/ Wkiol1uajzqtdd0uaafckflmrum299.jpg "longdesc="./1422006929911.png "height=" 272 "/>

650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1422007045762 "border=" 0 "alt=" 1422007045762 "src=" http://s3.51cto.com/wyfs02/M01/5B/21/wKioL1UAJzvCbFj_AAI_ Q2x5w-s826.jpg "longdesc="./1422007045762.png "height=" 322 "/>

650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1422007139216 "border=" 0 "alt=" 1422007139216 "src=" http://s3.51cto.com/wyfs02/M02/5B/21/ Wkiol1uajz2yqx6jaakg1mnoozw548.jpg "longdesc="./1422007139216.png "height=" 397 "/>

650) this.width=650; "style=" border-bottom:0px;border-left:0px;border-top:0px;border-right:0px; "title=" 1422007539361 "border=" 0 "alt=" 1422007539361 "src=" http://s3.51cto.com/wyfs02/M00/5B/21/ Wkiol1uajz3d6alxaaerxykun54878.jpg "longdesc="./1422007539361.png "height=" 315 "/>

EIGRP Comprehensive Lab Notes