Problem description

1. Customer OSN3500 device cross boards frequently report BUS_ERR alarm

Alarm information

BUS_ERR alarm and HSC_UNAVAIL are reported on 9-SXCSA boards.

2. Host version is 5.21.17.31.

Processing

1, first confirm the service, there is no interruption, and the 9-board position crossover board is the backup board.

2、Check the detailed alarm information of 9-SXCSA single board, first check the HSC_UNAVAIL alarm, its parameters are: 0x04 0x01 0x09 0xff 0xff

For the HSC_UNAVAIL alarm, HSC_UNAVAIL is only reported by the backup board, and the meaning of this alarm is as follows:

Parameter 1: Indicates the type of the alarm.

0x01 indicates that the state of this board is bad

0x02 indicates that a service board detects that this board is bad

0x04 indicates that this board detects type 2BUS_ERR

BIT[3 to 6]: reserved

0x80 indicates that the hard reset of the reserve board was successful for less than 5 minutes

Parameter 2: Indicates the primary and backup status of the unavailable single board.

0x00 indicates the main board

0x01 indicates the spare board

Parameter 3: Indicates the slot of the unavailable board .

Indicates the slot number of the unavailable board, the values are 0x09, 0x0A, 0x50, 0x51.

3. From the above, you can see that this alarm is generated because the SXCSA board at board position 9 has detected a type 2BUS_ERR alarm.

4, view BUS_ERR alarm, its parameters are: 0x0d 0x04 0x06 0x02 0xff and 0x0d 0x04 0x04 0x02 0xff

BUS_ERR alarm, the meaning of the alarm is as follows:

Parameter 1: Default indicates single board logical board position (including expansion board position).

If parameter 4 is 0x03, then parameter 1 indicates the cross chip number where the internal bus is located.

Parameter 2: Default indicates the bus serial number in this board position.

If parameter 3 is 0x03, then parameter 2 indicates the physical serial number of the internal bus in the chip.

Parameter 3: Different bits indicate different alarm presence status.

0x10 means that BUS_LOS exists.

0x08 indicates the presence of FIFO overflow

0x04 means there is B1 error

0x02 indicates the presence of BUS_OOF.

0x01 indicates the presence of BUS_OOA.

Parameter 4 : Indicates the alarm type of BUS_ERR.

0x01 indicates type I BUS_ERR, which is detected by a single cross board.

0x02 indicates type IIBUS_ERR, which is detected by two cross-boards handshaking.

0x03 indicates a Type IIIBUS_ERR, which is detected by the internal bus of the cross-board.

Parameter 5: Invalid parameter

5, as can be seen from above, it is due to the two cross board handshake detected that the 13 board bits generate a large number of error codes. Where:

0x0d means board 13, corresponding to the actual board SLQ16.

0x04 represents the 4th bus of the 13th board position, corresponding to the 4th optical port of the actual SLQ16

0x04: indicates the existence of B1 error code; 0x06 (where 06=02+04 ): indicates the existence of both 0x01OOF and 0x04B1 error code

Confirmed that it should be caused by the 13-SLQ16-4 optical port constantly generating a large number of B1 BERs and OOFs.

6, check the alarm information of 13-SLQ16, found that the 4-optical port keeps reporting R_LOF, confirmed by the customer that it is carried on the OTN/" target="_blank">wave division of other manufacturers, and at present the OTN/" target="_blank">wave division line is in the state of cutover.

7, to be OTN / WDM line stabilization, check the 13-SLQ16-4 optical port, R_LOF alarm disappears, the performance is no error code, again check the alarm of the cross-board, all alarms end. It is confirmed that the OTN/" target="_blank">WDM line cut causes our equipment to keep reporting R_LOF, which leads to alarms on the cross board.

Root cause

1. Failure of the backup cross board

2. Abnormal state of the main and backup crossover backups

3、Line error code

4、Main cross board failure

Recommendations and Summary

1, under normal circumstances OSN8800 line board reported R_LOF will not lead to OSN3500 cross board reported BUS_ERR alarm, from the failure to see, in other manufacturers OTN/" target="_blank">wave division equipment docking, friends of OTN/" target="_blank">wave division line is in an abnormal state, our docking line reported R_LOF , the line is in an abnormal state, our company docking line is not stopped. When the WDM equipment is docked, the line board of our company keeps reporting R_LOF, which indicates that the status detected by the cross board is good or bad, thus causing the cross board to report BUS_ERR alarm.

2、Description of "different bits indicate different alarm status" in alarm parameter:

The alarm parameter 0x is a fixed value, and the last two bits are counted in hexadecimal: so the 10 of 0x10 corresponds to the 16 of decimal.

Alarm parameter 0x is a fixed value, the last two digits are counted in bits of 2: so only 0x01, 0x02, 0x04, 0x08,0x10 (corresponding to 0, 1, 2, 3, 4 powers of 2 in that order) are interpreted.

Corresponding to other alarm parameters, they are calculated according to bit bits, and the calculation is as follows: 06 = 02 ( 1st power of 2 ) + 04 ( 2nd power of 2 ); 05 = 01 ( 0 power of 2 ) + 04 ( 2nd power of 2 );