Voltar para EIGRP – BALANCEAMENTO

Re: EIGRP – BALANCEAMENTO

Home Fórum Problemas do mundo real EIGRP – BALANCEAMENTO Re: EIGRP – BALANCEAMENTO

#55986

Vamos tentar novamente… 🙂 Minha configuração é:

routerA:

interface FastEthernet0/0
bandwidth 8192
ip load-sharing per-packet
ip address 172.16.0.1 255.255.255.252

interface ATM1/0.1 point-to-point
bandwidth 2048
ip load-sharing per-packet
ip address 10.10.10.2 255.255.255.0

router eigrp 1
variance 6
network 10.10.10.0 0.0.0.255
network 172.16.0.0 0.0.0.255
network 192.168.0.0
no auto-summary

A configuração do routerB é MUIIIITO semelhante.

Verificação:

routerA#sh ip route 192.168.80.1
Routing entry for 192.168.80.0/24
Known via “eigrp 1”, distance 90, metric 442880, type internal
Redistributing via eigrp 1
Last update from 10.10.10.1 on ATM1/0.1, 00:11:08 ago
Routing Descriptor Blocks:
* 172.16.0.2, from 172.16.0.2, 00:11:08 ago, via FastEthernet0/0
Route metric is 442880, traffic share count is 240
Total delay is 5100 microseconds, minimum bandwidth is 8196 Kbit
Reliability 255/255, minimum MTU 1500 bytes
Loading 1/255, Hops 1
10.10.10.1, from 10.10.10.1, 00:11:08 ago, via ATM1/0.1
Route metric is 1379840, traffic share count is 77
Total delay is 5080 microseconds, minimum bandwidth is 2048 Kbit
Reliability 255/255, minimum MTU 1514 bytes
Loading 1/255, Hops 1

routerA#sh ip cef 192.168.80.1 internal
192.168.80.0/24, version 40, epoch 0, per-packet sharing
0 packets, 0 bytes
via 172.16.0.2, FastEthernet0/0, 0 dependencies
traffic share 240, current path
next hop 172.16.0.2, FastEthernet0/0
valid adjacency
via 10.10.10.1, ATM1/0.1, 0 dependencies
traffic share 77
next hop 10.10.10.1, ATM1/0.1
valid adjacency

0 packets, 0 bytes switched through the prefix
tmstats: external 0 packets, 0 bytes
internal 0 packets, 0 bytes
Load distribution: 0 1 0 1 0 1 0 1 0 0 0 0 0 0 0 0 (refcount 1)

Hash OK Interface Address Packets
1 Y FastEthernet0/0 172.16.0.2 0
2 Y ATM1/0.1 point2point 0
3 Y FastEthernet0/0 172.16.0.2 0
4 Y ATM1/0.1 point2point 0
5 Y FastEthernet0/0 172.16.0.2 0
6 Y ATM1/0.1 point2point 0
7 Y FastEthernet0/0 172.16.0.2 0
8 Y ATM1/0.1 point2point 0
9 Y FastEthernet0/0 172.16.0.2 0
10 Y FastEthernet0/0 172.16.0.2 0
11 Y FastEthernet0/0 172.16.0.2 0
12 Y FastEthernet0/0 172.16.0.2 0
13 Y FastEthernet0/0 172.16.0.2 0
14 Y FastEthernet0/0 172.16.0.2 0
15 Y FastEthernet0/0 172.16.0.2 0
16 Y FastEthernet0/0 172.16.0.2 0
refcount 6

Agora os testes com PING:

routerA#ping 192.168.80.1 size 1000 repeat 1000 source 192.168.0.1

routerB#sh int | i proto|rate
FastEthernet0/0 is up, line protocol is up
Queueing strategy: fifo
30 second input rate 39000 bits/sec, 7 packets/sec
30 second output rate 39000 bits/sec, 7 packets/sec

ATM1/0 is up, line protocol is up
Queueing strategy: fifo
30 second input rate 14000 bits/sec, 3 packets/sec
30 second output rate 14000 bits/sec, 3 packets/sec

Algum tempo depois…

routerB#sh int | i proto|rate
FastEthernet0/0 is up, line protocol is up
Queueing strategy: fifo
30 second input rate 51000 bits/sec, 8 packets/sec
30 second output rate 51000 bits/sec, 8 packets/sec

ATM1/0 is up, line protocol is up
Queueing strategy: fifo
30 second input rate 18000 bits/sec, 3 packets/sec
30 second output rate 18000 bits/sec, 3 packets/sec

Minhas considerações:

1) Veja que o “traffic share count” no comando “show ip route” está representando proporcionalmente a capacidade dos dois links.

2) O resultado do comando “show ip cef internal” nos mostra como efetivamente os pacotes serão enviados através da linha “Load distribution: 0 1 0 1 0 1 0 1 0 0 0 0 0 0 0 0 (refcount 1)” e nas linhas subsequentes.

3) Sim, o cliente pode perceber MUITA lentidão por causa da “desordem” dos pacotes, principalmente se os links tiverem um tempo de resposta muito diferente do outro e/ou algum deles estiver com jitter muito grande.

4) Sobre a questão de vender sem fazer lab isso é muito comum :-).

5) Para entender o funcionamento do EIGRP com load-balance em caminhos desiguais, aconselho ler esse documento:
http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a008009437d.shtml

6) Se conselho fosse bom, eu vendia!!! :-D (essa é a melhor parte!!!! rs).

Boa sorte!

ps. Marco, agora que já escrevi tudo isso, acho que vale a pena escrever um artigo completo sobre isso, né?