Proliant DL360 G3 with 1611 error code

[George Clarkson]

Hi all,

I have searched within the forum for a possible solution, but it appears I couldn't fine anything that would apply to my case.

Problem is I bought a used HP PROLIANT DL360 G3 server in pristine conditions, all was perfectly fine. During OS installation, I had to reboot the server, and the post message appeared stating:

"1611 SYSTEM Zone fan assembly failure detected".
And then, as with every body else, the server shuts down after 5 seconds.

I did notice that the RPM cable of one of the 3 fans on the system fan board gives a strange voltage output: all fans have are connected to the system fan board via a 8 pin connector, of which only 7 pins are used. The fan module has 2 redundant fans, and the color code is the following:

first fan:
red - 12v
yellow - rpm
black - earth

second fan:
orange - 12v
white - rpm
grey (two cables soldered together) - earth

Fan is: Nidec model DR04XLG-12PUS1 03B
Fan assy is: 307525-001

I connected a multimeter to all the cables, to check voltages and eventual short circuits, and discovered that on one (and only one) of the fans, there's a voltage of 3,3v on the RPM cable. I double checked, powering down the server and powering it up again even after 24hrs, and the situation stays the same.

I swapped the fan with another one of the same fan assy, still getting the same 3,3v on the RPM line.

Removing the fan and powering up the server, and checking directly on the connector of the fan board, I can still get 3,3v from the respective pin. No other fan connector has such voltage on that pin.

I already ordered on ebay a used fan module, but I wonder why this happened. Is this issue prone to happen again, in some time? I do know I cannoct expect perfect functionality from a used server, but not seeing any burn signs, and the defect coming out of the blue, looks indeed strange. I already contacted the seller, and he assured me the server has been fully checked for 100% functionality prior to shipment. It is on of the biggest resellers here in Germany, so I do tend to believe them.

Also, is there a way to reset the error code in bios?

Any advice is welcome.

[LaneB]

Contrary to popular belief, not all problems can be diagnosed with a multimeter. ;-)

The tachometer pin on a typical RPM-sensing fan will produce a series of pulses rather than a steady DC voltage. The frequency of such pulses, or possibly the pulse width, will be determined by the fan's speed. You need an oscilloscope to view the tach output if you expect to get any useful result.

Your multimeter is probably doing what it can to average out the pulse train, and present it as a DC voltage. It should not be depended on for any sort of meaningful measurement.

Happy tweaking.

[George Clarkson]

[quote]
Contrary to popular belief, not all problems can be diagnosed with a multimeter. ;-)

The tachometer pin on a typical RPM-sensing fan will produce a series of pulses rather than a steady DC voltage. The frequency of such pulses, or possibly the pulse width, will be determined by the fan's speed. You need an oscilloscope to view the tach output if you expect to get any useful result.

Your multimeter is probably doing what it can to average out the pulse train, and present it as a DC voltage. It should not be depended on for any sort of meaningful measurement.

Happy tweaking.[/quote]

Hi LaneB, thank you for taking your time in replying to my question. I know a multimeter does not make miracles. And I know how a PWM signal works. But I never stumbled upon a 3.3v signal on a RPM cable, in more than 15years of worling in the IT business. Maybe I never searched hard enough... :-)

As you probalby already read in my initial post, I already ordered a "new" fan board, so the problem "should" be solved. Only if it does not come back again after some time. Which I would consider unusual. My post was rather to see if someone had already encountered similar issues with this particular model, and if these issues were effectively due to a malfunction of the fan board. In this case, a replacement should eliminate the problem at its root. But what if there is something else that causes the defect on the fan board? The power supply maybe? That was the sense of my question...

Anyway, I'll only know when I get the new-used fan board, later today. I'll update this post eventually if every is fine, or if it went bust.

Regards.

[George Clarkson]

Quick update on the issue:

I replaced the system fan board but the problem persisted, so I decided to do something I should have done before spending the money on the board: swap each fan one by one with the fans on the cpu fan board. I found one of the three fans on the system board, only one made the cpu fan board report error, so I assumed the problem was with the fan, and not with the board.

I immediately ordered a new fan module, which arrived a couple of days later. Installed it yesterday, and the server boots up perfectly fine. It was indeed the fan...

I am keeping the new fan board installed anyway, since I want to try the presumibly faulty fan with the new board. Hoping that it doesn't damage it.

Thanks for the help.

Closing this thread.