Hi dear friends.
Today, I will share a case about the analysis of OLT unreachable.
The product: MA5680T
Software version: MA5600 V800R018C10SPH221
Case Description
An MA5680T of the user is unreachable by the NMS. The cause needs to be analyzed.
Handling Procedure
1. Query faults on NMS. It is found that the MA5680T was unreachable twice on October 16
2. Check the data on the OLT. It is found that the OLT runs normally, the upstream port is not interrupted, the LACP is normal, and no abnormal alarm is reported. The LACP has not been intermittently disconnected since September 27.
- During the unreachable period on October 16, no alarm was reported.
- During the unreachable period, the OLT receives SNMP packets whose source IP addresses are not in the firewall list and the device discards the SNMP packets. This event does not cause the device to be unreachable.
- According to the preliminary analysis, the possibility that the uplink port is interrupted or LACP flapping causes the device to be unreachable to the NMS is excluded.
3. Check whether the device is unreachable by the NMS due to factors such as high CPU usage and packet flood. The check result shows that the management VLAN 9 of the device is transmitted upstream through the H801ETHB board in slot 7.
- Log in to the transparent channel of the H801ETHB board in slot 7, print ARP packets, and check whether the ARP packet flood occurs. The command output shows that the upstream port receives a large number of broadcast ARP packets from VLAN 15.
Check the configuration. It is found that a Layer 3 interface with vlanif 15 configured on the switch is in the Down state, and the VLAN is not added to any other interface.
Check the forwarding principle of the ETHB board. After vlanif 15 is configured on the device, the ETHB board captures ARP packets corresponding to vlanif 15 and sends them to the CPU for processing.
The cause is that the ETHB board receives a large number of ARP packets from vlanif 15. As a result, the ARP packets of management VLAN 9 cannot be processed normally. As a result, the ETHB board is temporarily unreachable from the NMS.
Root Cause
The management VLAN receives a large number of broadcast ARP packets on the ETHB upstream board, which affects normal management ARP interaction and causes temporary unreachability.
Solution
Delete the vlanif 15 that is not used from the device. Command: undo interface vlanif 15
Suggestions
Do not retain redundant Layer 3 interface data on the device.
Thanks.
Leave a comment