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

Traffic Is Intermittently Interrupted When CPOS Interfaces and E1 Interfaces Are Interconnected at a Site

Publication Date:  2019-04-22 Views:  18 Downloads:  0
Issue Description

Branch 1 of a company uses Huawei NE40E-X3s and is connected to the HQ through E1 lines.

According to the customer feedback, the ping to the HQ address fails and the physical status and protocol status of the local serial interface are both up. No abnormal alarms are generated on the NE40E-X3s.

Handling Process

1. Networking environment and simulated configuration:

Branch 1 (Serial2/0/0)--------(Serial7/0/0) HQ (Serial7/0/1)-----(Serial3/0/0) branch 2

Data configuration:

Simulate HQ, and configure the IP address 1.1.1.1 to connect to branch 1 and the IP address 2.2.2.1 to connect to branch 2.

Interface configuration:

Simulate branch 1, and configure the IP address 1.1.1.2 to connect to the HQ.

Simulate branch 2, and configure the IP address 1.1.1.2 to connect to the HQ.

2. Test result:

Ping branch 1 and branch 2 from HQ. The ping succeeds.


Ping HQ from branch 1. The ping fails.


Ping HQ from branch 2. The ping succeeds.



Mirror traffic on Serial 7/0/0 connecting the HQ and branch 1. The packet header information shows that the HQ does not send ICMP packets as a reply.


Mirror traffic on Serial 7/0/1 connecting the HQ and branch 2. Packets can be exchanged properly.

3. Cause analysis:

(1) Ping the HQ (1.1.1.1) from branch 1. The source IP address of the ICMP echo request packet is 1.1.1.2 and the destination IP address is 1.1.1.1.

(2) The PPP addresses of both branch 1 and branch 2 are 1.1.1.2. Therefore, equal-cost route are generated on the HQ.

(3) The packet header information shows that the HQ sends ICMP echo replay packets to branch 2.

(4) In this test, AR2240 is used to simulate the HQ, and the equal-cost routes are forwarded based on flows. Therefore, the ping on the HQ (1.1.1.1) from branch 1 fails but the ping on the HQ (1.1.1.1) from branch 2 succeeds.

(5) If the HQ device forwards equal-cost traffic based on flows, the ping to the HQ from branch 1 may fail for a while and succeed for a while.

(6) The IP addresses of both ends of a PPP link are on different network segments and the interfaces can go up after negotiation. Therefore, the HQ generates equal-cost routes although the IP address of branch 2 is incorrect.



Root Cause

When the network of branch 2 is adjusted, the IP address of branch 1 is used, causing equal-cost routes on the NE40E-X3.

Solution

Reconfigure an IP address for branch 2.

Suggestions

The service restoration takes a lot of efforts because the troubleshooting is performed on branches. Due to the special characteristics of PPP, you are advised to manually shut down the unwanted serial interface on the HQ. This prevents the problem caused by malicious configuration or misoperation.

END