The NE40Es, as the gateway of UMG, were configured with VRRP and BFD. VRRP heartbeat packets and BFD packets were forwarded by S65 switches.
BFD session flapping occasionally occurred on a subinterface of each NE40E.
Huawei completed the following steps to diagnose the problem:
1. Checked the BFD session flap log BFD/4/STACHG_TODWN on the master NE40E.
According to the Diagnostic parameter, the flap type was DetectDown, meaning that BFD packets were not received as expected during the BFD session detection period on the NE40E (BFD down).
2. Checked the BFD session flap log BFD/4/STACHG_TODWN on the backup NE40E. According to the Diagnostic parameter, the flap type was NeighborDown, meaning that the backup NE40E received the BFD packets from the master NE40E.
3. Suspected that the link between the NE40Es and UMG was uni-directionally available.
4. Queried data on the S65s. A large number of packets with a certain VLAN were discarded on an interface between the S65s.5. Migrated services to another interface on the S65s. The problem was solved.
When a similar problem occurs, checking BFD flapping logs is useful.There are four failure types: AdminDown, ReceiveAdminDown, DetectDown, and NeighborDown.