No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade
Knowledge Base

Packet Loss Occurs During OSPF Link Switching on an NE20E

Publication Date:  2019-05-13  |   Views:  257  |   Downloads:  0  |   Document ID:  EKB1100014930

Contents

Issue Description

Two Cisco 7606 routers and two Huawei NE20-Ss are connected in a square-shaped topology. The two Cisco devices and two Huawei devices both work in backup mode. OSPF runs between these devices. If the master Huawei device or master Cisco device becomes faulty, services are interrupted for about 40 seconds.

The network topology is as follows:


Handling Process

1. Check that the device configuration is correct.

  2. Ping the master 7606 router from the core switch. It is found packet loss occurs for about 40 seconds. Ping the master NE20E from the core switch. Packet loss occurs also for about 40 seconds.

  3. Disconnect the link with the master NE20E.. Check the OSPF routing table. It is found that the next hop is instantly switched to the slave NE20E.

  4. Confirm that OSPF convergence works properly on NE20Es.

  5. Perform the same operations on 7606 routers. It is found that OSPF convergence also works properly on 7606 routers.

  6. Disconnect the link connecting two 7606 routers and the link connecting two NE20Es. Then, ping these devices from the core switch. It is found that only four packets are lost, and services are switched quickly.

  7. Check that an encryption device exists on the links between 7606 routers and NE20Es. When the link on one end is disconnected, the interface of the master device on the link on the other end is still up. Therefore, OSPF convergence succeeds after Hello Dead (40s) times out.

Root Cause

Other devices exist between the links connecting the master devices. When the link on one end is disconnected, the interface of the master device on the link on the other end is still up. Therefore, OSPF convergence succeeds after Hello Dead (40s) times out.

Solution

Configure BFD and OSPF association with BFD on devices so that routers on both ends of a link can be quickly notified of link failures. This triggers fast OSPF convergence.