EIGRP Summarization
- EIGRP supports two types of summarizations: auto-summary, which is disabled by default, and manual summarization. Let's focus on manual summarization.
- If a Query is sent and no reply is received in 3 min the route will go into SIA (Stuck in active)
- We should avoid automatic summaries due to the potential outcomes they could produce. Also, this type of summarization does not work with IPv6.
- The best SID metric of the prefixes being summarized will be selected as the metric of the summarized route.
- Classic metric summarization
Classic metric summarization is configured at the interface level for both IPv4 and IPv6. The following is an example of how to set it up.
R3 has three loopback interfaces with consecutive IPv4 and IPv6 addresses, as illustrated below.
interface Loopback1
ip address 192.3.1.1 255.255.255.0
ipv6 address 2001:192:3:1::1/64
ipv6 eigrp 1
!
interface Loopback2
ip address 192.3.2.1 255.255.255.0
ipv6 address 2001:192:3:2::1/64
ipv6 eigrp 1
!
interface Loopback3
ip address 192.3.3.1 255.255.255.0
ipv6 address 2001:192:3:3::1/64
ipv6 eigrp 1
These subnets have been advertised to EIGRP. From the interface configuration, we can see that the interfaces were advertised in EIGRPv6. For the IPv4 subnets, they were advertised using the network command, as shown below.
router eigrp 1
network 192.3.0.0 0.0.255.255
R3 has two neighbors, R1 and R4. Since the summarization is performed on the interface where the neighbors are connected, both interfaces connecting to R1 and R4 will be set up to summarize the network, as illustrated below.
interface GigabitEthernet1
ip summary-address eigrp 1 192.3.0.0 255.255.0.0
ipv6 summary-address eigrp 1 2001:192:3::/48
interface GigabitEthernet2
ip summary-address eigrp 1 192.3.0.0 255.255.0.0
ipv6 summary-address eigrp 1 2001:192:3::/48
- Named mode summarization
R1 has two neighbors, R2 and R3. Like R3, R1 has three loopback interfaces with contiguous IPv4 and IPv6 addresses, as shown below.
interface Loopback1
ip address 192.1.1.1 255.255.255.0
ipv6 address 2001:192:1:1::1/64
!
interface Loopback2
ip address 192.1.2.1 255.255.255.0
ipv6 address 2001:192:1:2::1/64
!
interface Loopback3
ip address 192.1.3.1 255.255.255.0
ipv6 address 2001:192:1:3::1/64
R1 has two neighbors, hence the configuration in named mode should apply to the af-interface X mode. Note that this cannot be done under af-interface default.
router eigrp CORE
!
address-family ipv4 unicast autonomous-system 1
!
af-interface GigabitEthernet1
summary-address 192.1.0.0 255.255.0.0
!
af-interface GigabitEthernet2
summary-address 192.1.0.0 255.255.0.0
!
address-family ipv6 unicast autonomous-system 1
!
af-interface GigabitEthernet2
summary-address 2001:192:1::/48
!
af-interface GigabitEthernet1
summary-address 2001:192:1::/48
R1 will install a route to the summarized network pointing to Null0, for loop avoidance.
R1#sho ip route eigrp | in 192.1
D 192.1.0.0/16 is a summary, 04:18:09, Null0
R1#sh ipv6 route eigrp | be 192:1
D 2001:192:1::/48 [5/1280]
via Null0, directly connected
R1#sho ip route 192.1.0.0
Routing entry for 192.1.0.0/16, supernet
Known via "eigrp 1", distance 5, metric 1280, type internal
Redistributing via eigrp 1
Routing Descriptor Blocks:
* directly connected, via Null0
Route metric is 1280, traffic share count is 1
Total delay is 1 microseconds, minimum bandwidth is 8000000 Kbit
Reliability 255/255, minimum MTU 1514 bytes
Loading 1/255, Hops 0
R1#show ipv6 route 2001:192:1::/48
Routing entry for 2001:192:1::/48
Known via "eigrp 1", distance 5, metric 1280, type internal
Route count is 1/1, share count 0
Routing paths:
directly connected via Null0
Last updated 03:03:02 ago
As stated in the beginning, the best SID metric of the prefixes being summarized will be selected as the metric of the summarized route.
From R1, we can analyze the metric to 192.4.0.0/16
1 - R1's routing and topology table
R1#sho ip route | in 192.4
D 192.4.0.0/16 [90/271360] via 10.1.3.3, 00:00:07, GigabitEthernet2
P 192.4.0.0/16, 2 successors, FD is 34734080
via 10.1.2.2 (34734080/34078720), GigabitEthernet1, serno 220
via 10.1.3.3 (34734080/34078720), GigabitEthernet2, serno 217
2 - Interface values on R4 for the subnets being summarized.
Loopback1 is up, line protocol is up
Hardware is Loopback
Internet address is 192.4.1.1/24
MTU 1514 bytes, BW 8000000 Kbit/sec, DLY 500 usec,
Loopback2 is up, line protocol is up
Hardware is Loopback
Internet address is 192.4.2.1/24
MTU 1514 bytes, BW 8000000 Kbit/sec, DLY 500 usec,
Loopback3 is up, line protocol is up
Hardware is Loopback
Internet address is 192.4.3.1/24
MTU 1514 bytes, BW 8000000 Kbit/sec, DLY 500 usec,
3 - We will now change the DLY for interface Loopback3 and see the new SID metric for summarized metric.
R4#conf t
R4(config)#int lo3
R4(config-if)#delay 40
4 - Verification from R1
R1#sho ip route | in 192.4
D 192.4.0.0/16 [90/220160] via 10.1.3.3, 00:00:47, GigabitEthernet2
R1#sho ip eigrp topology
P 192.4.0.0/16, 2 successors, FD is 28180480
via 10.1.2.2 (28180480/27525120), GigabitEthernet1, serno 220
via 10.1.3.3 (28180480/27525120), GigabitEthernet2, serno 217
As we can see, the metric and FD are lower that before
In named mode, you can specify the metric using the K values or modify the AD.
R1#sho ip eigrp topology 192.4.0.0 255.255.0.0
EIGRP-IPv4 VR(CORE) Topology Entry for AS(1)/ID(1.1.1.1) for 192.4.0.0/16
State is Passive, Query origin flag is 1, 2 Successor(s), FD is 28180480, RIB is 220160
Descriptor Blocks:
10.1.2.2 (GigabitEthernet1), from 10.1.2.2, Send flag is 0x0
Composite metric is (28180480/27525120), route is Internal
Vector metric:
Minimum bandwidth is 1000000 Kbit
Total delay is 420000000 picoseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
Originating router is 4.4.4.4
10.1.3.3 (GigabitEthernet2), from 10.1.3.3, Send flag is 0x0
Composite metric is (28180480/27525120), route is Internal
Vector metric:
Minimum bandwidth is 1000000 Kbit
Total delay is 420000000 picoseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
Originating router is 4.4.4.4
We will change these values in R4 and see the result.
Below are the options of what we can set for the summarized routes.
R4(config-router-af-topology)#summary-metric 192.4.0.0/16 ?
<1-4294967295> Bandwidth metric in Kbits per second
community Community tag for summary route
distance Set administrative distance for summary route
R4(config-router-af-topology)#summary-metric 192.4.0.0/16 6 1000000 1 255 1 1500
As result R1 now sees different values in the topology and routing table.
R1#sho ip route | be 192.4
D 192.4.0.0/16 [90/20480] via 10.1.3.3, 00:01:47, GigabitEthernet2
[90/20480] via 10.1.2.2, 00:01:47, GigabitEthernet1
R1#sho ip eigrp topology 192.4.0.0 255.255.0.0
EIGRP-IPv4 VR(CORE) Topology Entry for AS(1)/ID(1.1.1.1) for 192.4.0.0/16
State is Passive, Query origin flag is 1, 2 Successor(s), FD is 2621440, RIB is 20480
Descriptor Blocks:
10.1.2.2 (GigabitEthernet1), from 10.1.2.2, Send flag is 0x0
Composite metric is (2621440/1966080), route is Internal
Vector metric:
Minimum bandwidth is 1000000 Kbit
Total delay is 30000000 picoseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
Originating router is 4.4.4.4
10.1.3.3 (GigabitEthernet2), from 10.1.3.3, Send flag is 0x0
Composite metric is (2621440/1966080), route is Internal
Vector metric:
Minimum bandwidth is 1000000 Kbit
Total delay is 30000000 picoseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
Originating router is 4.4.4.4
Leak-map
We ca use summarization with leak-map to allow some prefixes from the summary to be advertised along with the summary address. As an example, R4 is advertising a summary route of 192.4.0.0/16 with contains 192.4.1.1/24, 192.4.2.1/24 and 192.4.3.1/24.
We would like that R1 receives the summary route and some of the specific prefixes. 192.4.2.1/24 should be received via R2 and 192.4.3.1/24, from R3.
Below is the current route for the summary address 192.4.0.0/16 in R1.
R1#sho ip route | in 192.4
D 192.4.0.0/16 [90/20480] via 10.1.3.3, 04:56:56, GigabitEthernet2
The changes will be done at R4 so that leak-map can be used to allow the above requirement.
1 - Create prefix-list
R4(config)#ip prefix-list LEAK2 seq 5 permit 192.4.2.1/24
R4(config)#ip prefix-list LEAK3 seq 5 permit 192.4.3.1/24
2 - Create the route-map
R4(config)#route-map LEAK2 permit 10
R4(config-route-map)#match ip address prefix-list LEAK2
R4(config-route-map)#route-map LEAK3 permit 10
R4(config-route-map)#match ip address prefix-list LEAK3
3 - Configure the summary under the interface.
router eigrp CORE
!
address-family ipv4 unicast autonomous-system 1
af-interface GigabitEthernet1
summary-address 192.4.0.0 255.255.0.0 leak-map LEAK3
!
af-interface GigabitEthernet2
summary-address 192.4.0.0 255.255.0.0 leak-map LEAK2
4 - Verify from R1 that the prefixes are leaked as requested. As we can see, we achieved the desired solution.
R1#sho ip route | in 192.4
D 192.4.0.0/16 [90/20480] via 10.1.3.3, 04:59:18, GigabitEthernet2
D 192.4.2.0/24 [90/271360] via 10.1.2.2, 00:00:16, GigabitEthernet1
D 192.4.3.0/24 [90/220160] via 10.1.3.3, 00:00:05, GigabitEthernet2
EIGRP Queries are sent to multicast address 224.0.0.4. The Reply is sent via unicast.
More
R1(tcl)#foreach VAR {
+>(tcl)#192.2.3.1
+>(tcl)#192.2.2.1
+>(tcl)#} { ping $VAR }
Nenhum comentário:
Postar um comentário