Service Interruption Occurs at a Customer Office in Country A

Service interruption occurs at a customer office in country A because of incorrect configuration of cascading SNCP protection.

Fault Symptom

At a customer office in country A, one 10G service carried by OptiX OSN 8800 equipment was under cascading SNCP protection. After a fiber cut occurred on the line, SNCP protection switching was performed but the 10G service was still interrupted.

Network Topology

The figure above shows the NEs that the 10G service traverses and provides the following information:

1. SNCP protection was configured for the link between NE1 and NE2 and for the link between NE3 and NE4.
2. The link between NE2 and NE3 is not protected.
3. SNCP1 and SNCP3 are two west SNCP protection groups, while SNCP2 and SNCP4 are two east protection groups.

Cause Analysis

According to the customer feedback, the 10G client service was interrupted but no alarm was generated on the WDM side. Therefore, Huawei determined the following possible causes:

1. Some fiber connections were incorrect.
2. The client-side transmitter on the tributary board in an SNCP protection group was faulty.

Procedure

1. Queried the alarms and performance events of the two tributary boards. No abnormality was found. Therefore, Huawei suspected the fault was due to incorrect fiber connections. Considering that the service was interrupted after the SNCP protection switching, Huawei ruled out the possibility that a fault was present on the unprotected link.
2. Specified TTI to be set for the PM section of the tributary board at both ends to send TTI bytes using the two tributary boards. The tributary board in SNCP1 correctly received the TTI bytes from SNCP4 but the tributary board in SNCP4 did not receive any TTI bytes from SNCP1. This information indicates that the fault was unidirectional and was located on the link from SNCP1 to SNCP4.
3. Checked the SNCP protection link at both ends. Specifically, Huawei first turned on the PM non-intrusive monitoring function for the line boards in SNCP3 to monitor the signal coming from the west line boards. Then Huawei checked that the line boards in SNCP3 had not received the TTI bytes from the tributary board in SNCP1. This indicates that the fiber connection error was located between SNCP1 and SNCP2.
4. Considering that the service was running on the working channel before the fiber cut and on the protection channel after the fiber cut, Huawei determined that the fiber connections between the line boards on the protection channel were incorrect.
   After the boards that are incorrectly connected are identified, if the wavelengths on the two boards are different, you can determine that the board at the receiving site is incorrectly connected. Go to the receiving site and correct the fiber connections. If the wavelengths on the two boards are the same, you need to go to the transmitting and receiving sites to locate the fault.

Conclusion and Suggestion

For non-cascading SNCP protection, you can verify the correctness of the protection configurations by simulating a fiber cut on each of the working and protection channels. However, for cascading SNCP protection, you need to test the working and protection channels on each protected link to verify the correctness of the protection configurations. Bear this in mind to verify all SNCP protection groups on each SNCP-protected link. If the SNCP configurations on one link are incorrect, a service interruption is likely to occur when the network is in service.

More blog:

Warning of Failed Cascading on the MA5616 Equipped with the JDSU GE Optical Module

How to Control Board Failure After an MxU Service Port Is Bound to a Rate Unlimited

What’s the IGMP Framework on the Device