Fiber cut between two nodes resulted in service interruption in an MSP ring, though there was no SF or SD condition in any other section of the ring. APS_INDI alarm appeared along with R_LOS alarm on one node but the second adjacent node didn't show APS alarm. Force switching was done but the problem didn't resolve. When we tried to get MSP status from T2000, attributes of one node (ID:1498) couldn't be queried. Using 'cfg-get-mspstate' command didn't returned any output. Software version of the node was V100R002C02B025SP01.
R_LOS, APS_INDI, TU_AIS and AU_AIS
We downloaded the configuration data from NMS to SCC and the problem was resolved. We were able to query the protection attribute and protocol status. Protection swithcing was tested by breaking fiber and it occured successfully.
As a permanent solution, Upgrade to V100R003C02B032SP01 version which contains the solution of the 24 bytes memory block OVR problem.
There was no MSP swithcing event found in this node. The problem was due to the 24 bytes memory block OVR problem. This overwrote the class’s v-ptr, arose the SCC reset on saving the database and finally cause the configuration database abnormal. After that some of the configurations like enable MSP protocol didn’t send to XC Board.
Protection switching test should be conducted on monthly / quarterly basis to ensure the correct status of the protocol. Protection attributes and protocol should be queried on regular basis and if any abnormality is found, it should be handled proactively.