Hello everyone!
I will share a case about intra-board 1+1 troubleshooting.
Description
A customer uses intra-board 1+1 protection for a 100G service (site A-site B). When a fiber cut occurs on the working path, the service is also interrupted. After the fiber is repaired, the service is restored. The customer asked why the service was interrupted when the protection channel was normal.
Handling Process
Step 1: Analyze the alarms, performance events, and switchover logs generated during service interruption.
1. When services are interrupted, the N402 board has receive optical power, but the port reports an R_LOS alarm.
2. When the user reports a service interruption, the OAU1 and OSC boards on the FIU side of site A on the working trail have no receive optical power, but the receive optical power at site B is normal. This indicates that the fiber at site A is cut. The receive and transmit optical power at both ends of the protection trail is normal.
3. No intra-board 1+1 protection switching event was found in the switching log, indicating that no switching occurred.
4. Use the signal flow analysis method to locate the fault. Checked the OAU1 board on the FIU side. It was found that no receive optical power was received when the fault occurred. Checked the second downstream OAU1 board and found that the receive optical power was 2 dBm lower than the normal value when the fault occurred. Checked the receive optical power of the working DCP board and found that the receive optical power was 2 dBm lower than the normal value when the fault occurred.
Step 2: Check the device configuration.
1. Checked the upstream board of the second OAU1 board. The board was a WSMD9 board, and found that there were wavelengths from other dimensions.
2. Checked the intra-board 1+1 protection configuration of the DCP board. It was found that the switching condition was " Variance Threshold Between Primary and Secondary Input Optical Power is 5dB" and no extended detection board was configured.
Step 3: Analysis
1. When the fault occurs, the fiber from site A to site B on the working trail is cut.
2. The WSMD9 board receives wavelengths from other dimensions, the WSMD9 and TD20 boards both transmit broadcast optical signals. Therefore, when a fiber cut occurs, the receive optical power of the second OAU1 and DCP board decreases by only 2 dB.
3. The intra-board 1+1 protection switching condition is " Variance Threshold Between Primary and Secondary Input Optical Power is 5dB ". The difference is only 2 dB. Therefore, the optical power of the working path is always received without switching. At this time, the receive optical power of the working trail has lost the lambda of site B. Therefore, although the N402 board received optical signals, the R_LOS alarm was reported and services were interrupted.
4. If the N402 board is configured as the extended detection board in the intra-board 1+1 protection, the 1+1 protection switching is normal and services are not interrupted.
Root Cause
In the colorless architecture, the extended detection board is not configured for intra-board 1+1 protection.
Solution
Set the N board to be protected as an extended detection board.
That's all, I welcome everyone to leave a message and exchange in the comment area!
Thank you！
