PSU Failure Follow
Procedure:
Please boot up to the ACCTON-DIAG for troubleshooting.
1.1 Check the system LED light of the PSU.
The following figure is the system LED light location for AS5916-54XL, AS5916-54XKS.
The following figure is the system LED light location for AS7816-64X.
The following figure is the LED light definition of the system PSU for AS5916-54XL, AS59154XKS, AS781664X.
Caution: This is the LED light definition of ACCTON-DIAG.
The LED behavior may be different on other NOS.
Please consult the NOS vendor.
1.2 Check the LED status on the PSU module power injection.
The following figure is the LED light definition of the AC/DC PSU modules.
PSU status | DC PSU | AC PSU |
PSU operates normally | Blue | Green |
PSU is failure | Red | Red |
PSU warning | Blinking Red/Blue | Blinking Red/Green |
PSU plugged into the switch but not power on | OFF | OFF |
NOTE, * Blinking frequency: red and green/blue on 1 sec and off 1 sec separately.
If AC LED of the PSU module show red, blink red, blink red/green.
Or if DC LED of the PSU module show red, blink red, blink red/blue.
This is the PSU module that has a failure, please return the unit back for RMA repair service directly.
1.3 Identify the PSU number.
The following figure is an AS5916-54XL, AS5916-54XKS schematic diagram of PSU1 and PSU2.
The following figure is an AS7816-64X schematic diagram of PSU1 and PSU2.
1.4 Execute pwrmod detect PSU status on ACCTON-DIAG.
Please connect power cords(power injection) to two PSU modules.
Otherwise, the test result is failed as expected.
Example: PASS on AS5916-54XKS and AS5916-54XL.
root@(none):/# pwrmod
Power Module Test ..............
Unit 0:
Vendor: 3Y POWER
Part Number: YM-2851FC01R
Serial Number: SA000X131925000681
MCU FWID: SPRIN851AMP3C306A00
MCU Firmware Version: A00
Fan: F2B
PSU0 present: Yes
PSU0 AC: Normal
PSU0 12V: Normal
PSU0 PG signal, PWOK-1: Low
PSU0 fan speed: 6800 RPM
Unit 1:
Vendor: 3Y POWER
Part Number: YM-2851FC01R
Serial Number: SA000X131925000682
MCU FWID: SPRIN851AMP3C306A00
MCU Firmware Version: A00
Fan: F2B
PSU1 present: Yes
PSU1 AC: Normal
PSU1 12V: Normal
PSU1 PG signal, PWOK-1: Low
PSU1 fan speed: 6296 RPM
Power Module Test: PASS
Example: FAIL on AS5916-54XKS and AS5916-54XL.
root@(none):/# pwrmod
Power Module Test ..............
Unit 0:
Error: PSU0, failed to get 'Vendor Name'
PSU0 present: Yes
PSU0 AC: Fail
PSU0 12V: Fail
Error: PSU0, failed to get status word.
Error: PSU0, failed to get fan speed.
Unit 1:
Vendor: 3Y POWER
Part Number: YM-2851FCR
Serial Number: SA070U461823010072
MCU FWID: SPRIN851AMP3C300A02
MCU Firmware Version: A02
Fan: NA
PSU1 present: Yes
PSU1 AC: Normal
PSU1 12V: Normal
PSU1 PG signal, PWOK-1: Low
PSU1 fan speed: 6800 RPM
Power Module Test: FAIL
Example: PASS on AS7816-64X.
root@(none):/# pwrmod
Power Module Test ..............
PSU 0:
Vendor Name: 3Y POWER
Part Number: YM-2851FD01R
Serial Number: TA020X142014000454
Direction: B2F(C3h:0x42 0x32 0x46 )
MCU FWID: SPRIN851AMP3C307A00
MCU Firmware Version: A00
VOUT: 11.921V(0xd2fb)
IOUT: 10.531A(0xd2a2)
POUT: 125.000W(0xf1f4)
TEMP1: +33'C(0x0021)
TEMP2: +46'C(0x002e)
TEMP3: +37'C(0x0025)
FAN_SPEED: 7000 rpm(0x1b6b)
CPLD1(0x3)=0x3c
PSU0 Present: Yes
PSU0 AC: Normal
PSU0 12V: Normal
PSU 1:
Vendor Name: 3Y POWER
Part Number: YM-2851FD01R
Serial Number: TA020X142014000456
Direction: B2F(C3h:0x42 0x32 0x46 )
MCU FWID: SPRIN851AMP3C307A00
MCU Firmware Version: A00
VOUT: 12.109V(0xd307)
IOUT: 12.234A(0xd30f)
POUT: 148.000W(0xf250)
TEMP1: +40'C(0x0028)
TEMP2: +45'C(0x002d)
TEMP3: +36'C(0x0024)
FAN_SPEED: 6600 rpm(0x1b39)
CPLD1(0x3)=0x3c
PSU1 Present: Yes
PSU1 AC: Normal
PSU1 12V: Normal
Power Module Test: PASS
All the above, if the Power Module Test result is FAIL, please return the unit back for RMA repair service.
2.1 Get continuous high/low value in power event message from BMC via "ipmitool sel list".
Model: AS5916-54XKS or AS5916-54XL
This is a known issue in BMC version 0.51 and previous version. The Vout value is read from PSU by BMC, this value is not stable since lack of inspection mechanism.
Eventually, we enable PSU PEC(Packet Error Check) to determine PSU return data is correct or not. In a normal state,
once it found that the current value exceeds/falls over the threshold(Vout <10, or > 13), an event log is generated.
The following is an irregular high/low error log.
Resolution: Update the BMC version to 0.52 or above. How to update the BMC via web management?
After updating the BMC version to 0.52 or above, if you find a power event log is generated, please use the Accton_diag to check the PSU and also submit a ticket to Edgecore support.
Comments
0 comments
Please sign in to leave a comment.