Description of the problem

On a Huawei OSN350 device, multiplexed segment ring protection is configured on two optical ports on a line board such as SLQ16. One day, the two multiplexed segments simultaneously experienced a multiplexed segment protection inversion, and the service was normal during the inversion, but after waiting for the recovery time, only one multiplexed segment recovered to the Idle state normally, and the other MSP protection group was always in the Switch state.

Alarm Information

MS_APS_INDI_EX

Processing

After the customer deactivates the MSP protocol for the whole network, the service is resumed.

Root Cause

As shown in the diagram inside the attachment:

Figure 1 shows that when the new k-interrupt for optical port 3 is processed, a new k-interrupt for optical port 1 is generated and then this interrupt is processed. Under normal flow, there is no problem.

Figure 2 shows that when the new k interrupt of optical port 3 is being processed, the interrupt register has not yet been read, and the new k of optical port 1 is generated, no interrupt will be generated at this time, but the new k interrupt status bit of optical port 1 (each multiplexed segment of each optical port occupies one bit to indicate the interrupt status) will be set. Therefore the new k for both optical ports at this time is capable of being processed.

Figure 3 shows that when the new k interrupt of optical port 3 is processed, after the command to read the new k interrupt is issued, it takes a period of time to pull the level high when reading the clear, and at this time, if the new k of optical port 1 comes, the FPGA will bottom out the level, but the cpu does not detect the falling edge, and will not generate a new k interrupt. At this time, the new k of the optical port 1 is lost.

Suggestion and Summary

For multiple optical ports above the multi-optical board are configured with multiplexing segment protection, if multiplexing segment protection occurs at the same time in these multiple optical ports, and then multiplexing segment protection recovery is carried out at the same time, there is a high chance (more than 50% chance in lab test) that the K bytes are lost in the recovery of multiplexing segments, and the state of the multiplexing segment protocol is in error (one end is in the Idel state, and one end is in the Switch state), and the service is interrupted. Restart the multiplexing segment protocol for the whole network of the abnormal multiplexing segment, the K bytes will be reported normally again, and the service will return to normal.

It is recommended that the existing network try not to configure multiple multiplexed segments protection on top of multiple optical boards at the same time (the R8C02SPC200 version and later releases have solved this problem, but all the releases before that time have this problem); if you have to do such configuration, it is recommended to set the inversion waiting recovery time of different multiplexed segments to be inconsistent, which can also avoid solving this problem.

The related technical information in this chapter and the troubleshooting process of SDH equipment are collected and organized by Shenzhen Optical Transmission Network Technology Company Limited ( www.opticaltrans.com), please retain! Our company specializes in Huawei SDH optical transmission equipment, SDH transmission equipment sales.