Hands-on: HSRP hot backup routing protocol experiment Summary

Without a layer-3 switch, I used two 2514 + 1924 switches + my ADSL kitten for an experiment.

Environment connection:

In VLAN 1924, two VLANs are divided. One is the network segment 192.168.0.0/24 and the other is the network segment 10.0.0.0/24. In the VLAN 192, the network segment is connected to my laptop NIC/The E0 ports of two 2514 routers. VLAN 10 is used to connect two 2514 E1 ports and the cat's Ethernet ports.

Set the laptop Nic gateway to any 2514 E0 port address, so that the computer can communicate with the outside world through a single router. Because my cat does not support static routing settings, as a result, after I set the ip route 0.0.0.0 0.0.0.0 10.0.0.138 (this IP address is my cat's Ethernet port address) on the vro, the computer (ip192.168.0.1) cannot communicate with this 138 address, it cannot communicate with the public network, because we all know that data packets from 192.168.0.0 can be sent to the cat, but no route table on the cat returns to the 192 network, A cat is a dead cat and can only automatically communicate with 10 networks, So I performed the following NAT translation on the vro:

ip nat inside source list 1 interface Ethernet1 overloadip http serverip classlessip route 0.0.0.0 0.0.0.0 10.0.0.138! ! access-list 1 permit 192.168.0.0 0.0.0.255

In this way, the vro converts 192 of the network data to 10 of the network, and configures the two routers. After the configuration, the computer can access the QQ website.

The same method ensures that two routers can be connected to INERNET separately.

(By the way, I encountered a strange problem during the above process. The computer can PING 1924 OF THE MANAGEMENT IP address, 1924 can also PING all routers, and the reverse path can also ping, however, when you ping a vroping from a computer, the vroping cannot PING the computer, and the router is so depressed for a long time. Check that the MAC address table of the vswitch does not match the MAC address and PORT of the current computer. later, I found out that I bound my computer Nic MAC to an interface of this switch in my experiment yesterday, and today I inserted it to another interface to delete the static definition of yesterday and solve the problem .)

Configure STANDBY as follows

Configure R2 first

Standby 1 ip 192.168.0.254

Standby 1 priority 90

Here the priority is deliberately reduced

Configure R1

Standby 1 ip 192.168.0.254

Check show standby after configuration. As expected, R2 becomes active because R1 has a higher priority, but R1 is configured later.

In this case, the preempt preemption parameter is not configured. Disable the E0 port on R2. During the debugging process, R1 becomes active. start the E0 of R2 again, and the activity is still R1, as expected. because it is not preemptible. in this case, if you want to preemptible A vro with a lower priority, will you be able "? The actual experiment results show that, even if the preemption is set, it is not obtained because of its low priority.

If no preemptible is configured for another test, lower priority is used as the activity, and higher priority is enabled to test whether the higher priority can be automatically taken over. The actual result is that if no preemptible is configured, even if the priority is high, the initiative cannot be obtained.

Experiments show that automatic preemption can be achieved only when priority is high + priority is configured on a high-priority router.

Interface tracking test:

When interface tracing is not enabled, the E1 line on the active vro is unplugged. The vro cannot identify a fault in the network line and considers itself active. As a result, the network cannot access the Internet. re-plug in the network cable, the network returns to normal, there is no router switch.

Enable API tracking:

Configure standby 1 track e 1 20 configure 20 floating priority units

The experimental results show that only the preemptible settings can be used properly. For example, R2 has a priority of 90, R1 is 100, and R1 is currently active. After the E1 line of R1 is removed, although the debugging information shows that the priority sent by R1.

At this time, after R1 is configured with preemptible, the connection to R1 is restored, and R1 becomes active again, taking over R1.

(T113)