On an OptiX OSN 7500 NE, the T2SL64s in slots 28, 29, and 39 report the POWER_ABNORMAL alarm with parameters 0x01 0x00 0x01 0x00 0x05 at intervals of three to five minutes. The NE version is 188.8.131.52. The board version is 4.13.
1. Check the power supply to the equipment. It is found that the power voltage is properly ?55.3 V. Check the boards for bent pins. It is found that the boards do not have bent pins.
2. Move the boards to lower slots. No alarm is reported. Insert other boards (at the same and different rates) to the upper slots. The alarm is reported. Therefore, the boards are functional.
3. During the alarm, query the boards by using the command optp:1c/1d/1e,0,9b,1,35,01,0. It is found that the active and standby 3.6 V power supplies output 1.3 V and 0.5 V voltages. After the alarm, query the boards. It is found that the active and standby 3.6 V power supplies work properly. Engineers suspect that the ?48 V power supply has a problem. Test the ?48 V power supply in the three slots on the backplane. It is found that the ?48 V power supply properly outputs a ?55 V voltage. The test methods are not discussed. For details, see the relevant documents.
4. The alarm parameters indicate that the boards use the standby 3.3 V power supply from the AUX. During the alarm, remove the AUX. The SL64s in slots 28, 29, and 30 are powered off. Therefore, the alarm reports are true.
5. Power off and then restart the equipment. The fault persists. Engineers attempt to replace the subrack. Before the attempt, it comes to the engineers that the upper part of the equipment has a high temperature. Query the board temperature. It is found that the board temperature is 53°C. Clean the air filter. Cool the equipment by using an electric fan. The alarm disappears. The alarm is not reported in the next days.
Therefore, the fault occurs because the equipment temperature is too high. Due to the high temperature, the power modules on the boards work poorly. As a result, the boards report the POWER_ABNORMAL alarm at intervals.
1. The external power supply fails. The boards have bent pins.
2. The T2SL64s fail. The boards, however, are unlikely to all fail.
3. The board software has an error and gives false reports.
4. The backplane has poor contact.
Before you solve a fault by replacing hardware, ensure that you find the root cause of the fault. In addition, remind customers of routine maintenance to prevent such rare problems.