As shown in the figure below, A-B-C-D-E-F-G-H-I-J-K-L-M-A forms a 2.5G SNCP ring, with A as the center site and Ethernet services between all sites and A. Site A is configured for EPLAN, and the other sites are configured for EPL, and on a certain day, the fiber-optic cable from site A to site M is interrupted, and the services are normally reversed, and the fiber-optic cable fault is repaired for 10 minutes. Ten minutes after the cable failure is repaired, the VCTRUNK from station A to station M reports a VCAT_LOA alarm, and the Ethernet service from station M to station A is interrupted.

VCAT_LOA

1. VCAT_LOA alarm is reported on VCTRUNK of EFS0 board, which indicates that the configured VCTRUNK time slots have passed through physical links with different distances, and the SDH paths (5 VC12s) of the bound time slots from station A to station M are viewed, and the paths are all the shortest paths, A-M, with no problems. 2.
2. Before the cable was interrupted, the service was normal, and after the cable was interrupted, the service could be reversed normally, basically ruling out the possibility of single board failure.
3. 10 minutes after the cable repair, VCAT_LOA alarm is generated, just after the recovery of SNCP switching, it may be part of the channel SNCP switching is not restored, check the abnormal events, and it is true that two of the five VC12s are still in the SNCP switching state, and in the SNCP service control menu, check the configuration of these two VC12s for the restoration of the non-recovery.

Two of the five VC12 time slots bound to VCTRUNK have SNCP inversion recovery mode configured as not recovered, and these two time slots are not inverted back to the working path after SNCP inversion recovery, resulting in the five time slots passing through physical links with different distances, while the protection path passes through longer distances and more network elements, and the delay time exceeds the time allowed by the virtual cascade delay, resulting in service interruption.

Under the SNCP service control menu, change the SNCP inversion recovery method of these 2 channels to recovery, the service is inverted to the working path, and the service returns to normal.

When configuring SDH services, try to ensure that the inversion recovery method of SNCP services at both ends is the same, so as to avoid sending and receiving paths taking different routes. When configuring Ethernet services, make sure that the physical links passing through the configured time slots are the same, so as to avoid the service interruption caused by too much delay of the virtual cascade.