Shulou(Shulou.com)06/01 Report--

Objective: to test the operation and processing mechanism of MASTER and SLAVE after IRF splitting.

1. Run OSPF between Core and IRF devices. The backbone area is area 0, and PC1,PC2 to CORE1.1.1.1 can be reached.

2. IP address table

Device port / VLANIP address COREloopback1.1.1.1/32

Interface vlan 2100 (interconnect) 10.24.61.143/24IRF access vlan 10 (user vlan) 10.1.1.254Universe 24

Interface vlan 2100 (interconnection) 10.24.61.142/24BFD_MADvlan 3213 (separate private network address)

PC110.1.1.1/24

1.1.1.1 of PC users to CORE is reachable.

3. PC1 long ping 1.1.1.1

、

4. Manual IRF splitting, while observing that there is no change in the ping packet on the PC1, and the performance of the MASTER device is good when the IRF is split. And the MAD port down dropped the prompt.

5. The MASTER equipment operates normally and is not affected in any way. Then I switch to the standby machine to observe, BFD port down, the whole machine has no ARP information.

There is no ARP information indicating that the device is not working and the device will not affect our existing network. This is the key point of the test.

6. Manually restore the BFD_MAD connection, restore the IRF, merge the main and standby, and restore all IRF links to normal. Restart the standby and rejoin the IRF group. After a brief outage of PC2, it began to resume business.

7. IRF fission test results

The results show that when the IRF is split, the MASTER service works normally and does not affect the service. Prepare the opportunity for the whole machine to go into dormancy and not work, so as to avoid repeated information transmission affecting the network, if affecting the production network is a particularly troublesome thing. This is what no one wants to happen!

In addition, someone will ask me if you are idle to test this. It's not an idle pain. I'm not relieved. When you do more than a variety of problems encountered, it is difficult to reassure people, or under their own hands the most at ease. All digital devices are actually the same, some time ago cut off Cisco's 4507R+E, the new version of IOS-XE. After splicing, when your ping latency is very small, the 10m bandwidth download speed reaches 1.5m. But the Internet is a problem, the Internet simply can not open the web page, change the MTU value, Cisco 4507 is the default MTU value, can not be changed. With a laptop directly connected to the Cisco core uplink firewall, the speed is fast and no problem, not the firewall problem. As soon as I had experience, my colleagues told me that 4507 of the second and third floors were checked, but the results were still not good. Only doubt the version, and finally the general generation gave a so-called minimum BUG IOS, and the problem was solved. When the problem cannot be found, you can only change the version. The problem that can end in half a day, twists and turns for a day and a half, and then the technology of the boss of the Cisco industry is very mature, but when you encounter such a problem, you say that the egg hurts or not. Nothing is worth it. We can believe it!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.