6850E fictitious power supply

[3rdm4r]

Hey Folks!

I have a weird behaviour with a 6850E-P24 (OS6850E-P24 6.4.4.645.R01, one power supply "Primary" (510W).
The switch "think" he has another power supply - which dont exist.

Output from the log...

TUE JAN 12 16:44:54 2016 LANPOWER info Power Supply #1 Present(0x40f)!
TUE JAN 12 16:44:54 2016 LANPOWER info 48v Present(0x6)!
TUE JAN 12 16:44:54 2016 LANPOWER info Power Supply #3 Bad!!(0x40f)!
TUE JAN 12 16:44:55 2016 LANPOWER info Power Supply #1 Up(0x40f)!
TUE JAN 12 16:44:55 2016 LANPOWER info Power Supply #3 Bad!!(0x40f)!



I try other power supplys, the same failure - the BPS-LED (backup PS) on the front of the chassie lights orange.
Knows someone these behaviour?

Thanks in advance!

[devnull]

Never had that.
"Have you tried turning it off and on again?"
https://www.youtube.com/watch?v=nn2FB1P_Mn8

If you can: Power-cycle and/or update the firmware. If the problem persists open a TAC case

[3rdm4r]

"No I have not tried that" - thank you for the hint

The firmware running on other 300 devices (same hardware), without these behaviour.
What do you mean with "power cycle" - is there an option to check the power supply or internal hardware-area from the switch?
I would like to try everything, before I opened a ticket. Thanks!

[devnull]

Power-Cylcle: unplug power-cable, wait some seconds, reconnect it.
Sometimes a reload has not fixed all the problems, whlist unplugging did.

Firmware: i know that. I always tell myself, it is just software - and there may be a glitch, some sun-radiation flipping a bit whatever.. therefore it may not affect other devices.

What i also experienced: (not with powersupplys) Linecards in Chassis switches are not completeley hot-plugable:
A empty slot may take any card, but a occupied slot can only be exchanged with a similar card.
I had issues with a card being removed years ago (switch uptime >1000 days)- nobody rememberd that there was a 24-port Gig card in the past - now the new 6*10 card did not work.

Worked fine after reboot..

Therefore: a reboot may fix plenty of issues!

[3rdm4r]

The decive lies expanded on my desk - i think there a enough power cycles in the past
Ok thanks for the replys, i think it will be a ticket for the TAC.

Should I get more information, you will read it here!

[kabalin]

hello.
have alcatel 6800
the same problem
how have You solved it?

[kabalin]

The decive lies expanded on my desk - i think there a enough power cycles in the past
Ok thanks for the replys, i think it will be a ticket for the TAC.

Should I get more information, you will read it here!

hello.
have alcatel 6800
the same problem
how have You solved it?

[3rdm4r]

Hey kabalin, sorry for the late response... I have so far found no time to experiment with the problem.
Before I contact the TAC, i would be play with different aos versions and other power-supplys of the device, to delimit the issue.

Do you have new insights ?

Greetz

[kabalin]

Hey kabalin, sorry for the late response... I have so far found no time to experiment with the problem.
Before I contact the TAC, i would be play with different aos versions and other power-supplys of the device, to delimit the issue.

Do you have new insights ?

Greetz

sorry no ideas

[3rdm4r]

Hey Folks!,

today I do a CPU-board reset on the said switch from the console (interrupt boot Operation, => reset), switch boot up and has recognize his stacking neighbors - the same issue.



During the boot process of the switch, appears the following message:



Not really an explanation or a reason for this issue, but after an hardware exchange (PS-900AC-P) is this fixed.

Greetings from germany!