During the Software Upgrade of an OSN 3500, from 18.104.22.168 to 22.214.171.124 version, a N2EGS2 board needs a Hard Reset in order to apply the new software version, but after the restoration of the board the configured EPL service doesn't work and the Trunk Link shows an alarm.
Before the Maintenance the services was OK.
The Trunk Link shows "ALM_GFP_dLFD" alarm. This Major alarm indicates GFP loss of frame delineation or the GFP frames are detected out of synchronization. During data transmission, failure in GFP framing causes service interruption.
As the alarm appeared after upgrade the software of the board, is necessary to check if the actualization was successful. The version that the board should have is 126.96.36.199, using the command "cfg-get-bdverinfo:bid", we checked the version of the N2EGS2 board, and it was right.
FPGA: (U5)310, (U41)210, (U42)210
BIOS: "2.14 - 2.19"
So, we have to check the state and the configuration of the service, as follow:
1. View the ALM_GFP_dLFD alarm parameters on the T2000, and confirm the relevant board and VCTRUNK.
2. On the T2000, we check that the number of uplink or downlink timeslots bound with the VCTRUNK at the local station is consistent with the number of downlink or uplink timeslots bound with the corresponding VCTRUNK at the opposite station. If not, increase or reduce the timeslots to make the number of the timeslots at both ends consistent.
3. As the alarm persists, we check whether there are the bit error alarms such as BIP_EXC, BIP_SD, B3_EXC and B3_SD in the timeslots bound with the corresponding VCTRUNK.
4. Finally we check the state of the SDH trails and if they had some alarms, but the SDH service was normal and without alarms.
5. Like the service has "ALM_GFP_dLFD" alarm and the data transmition keep failures, we deactivate in the follow order t"he EPL ethernet service, the Trunk link and the SDH service", and activate again in the follow order "the SDH service, the Trunk link and the EPL ethernet service" and the service was reestablished without any alarm.
The operation of deactivate and activate again the service restored the lost synchronization of the GFP frames when the N2EGS2 board was reset
As the alarm appeared after the software upgrade is necessary check the software version again, afterwards, check the possible causes.
1. The number of uplink or downlink timeslots bound with the VCTRUNK at the local station is inconsistent with the number of downlink or uplink timeslots bound with the corresponding VCTRUNK at the opposite station.
2. Bit errors are present in the timeslots bound with the VCTRUNK.
3. The SDH services are in normal state.
Please be aware that the services are OK before any maintenance window, or pay attention to some happened event during the activity.
Take in mind save the configuration of the Ethernet services like preventive.