Proliant ML310e Gen8 making a LOT of noise!

[mattm591]

I recently purchased a Proliant ML310e Gen8 to replace a dying Proliant ML110G5.

Several reviews of the server commented on how quiet it was, which is a plus for me as the server sits with me in the office!

This new server makes a RACKET! It's considerably louder than the ML110G5, the fans sound like they're constantly going at least 50%, despite it being quite cool in the office and the server load being at 0.

The noise starts from first boot and has now continued for several solid days (tried a few reboots initaly to see if it was teething problems, never helped).

It's running Debian 6, and has a 3ware PCI-X raid card installed. I noticed a similar problem with a different Gen8 server that might have been caused by using a non-standard PCI card, and wondered if this was part of the problem?

Does anyone have any ideas as to possible solutions?

P.S. this thread has been moved from from Servers > General to ProLiant Servers (ML,DL,SL). - Hp Forum Moderator

[Suman_1978]

Hi,

What is the BIOS and iLO 4 firmware versions on your Gen8 server?

iLO 4 firmware has some fixes for fan speed, do check the Revision History of iLO4.

Checkout the BIOS and iLO4 firmwares here.

PCI card may or may not cause fan speed.

Thank You!
I am a HP employee
__________________________________________
Was the post useful? Click on the white KUDOS! Star.

[mattm591]

Thanks for the response.

iLo firmware is 1.22. Checked and updated everything that can be updated.

Still running with the same 50% fan usage constantly.

[TVVJ]

Reseat the card and check if the issue is resolved. Also, reseat components such as other expansion cards, memory modules.

Ensure that the third party card is of the correct dimension and  PCIe specification for the slot in which it is being installed.

Refer to page 70 of the server’s maintenance and service guide that details the speed of the PCIe slots depending on the processor installed and ensure that the third party card is in compliant thereof. Click here to access the server’s maintenance and service guide.

Try clearing the server NVRAM. To do so the following steps may be performed:

1. 1.      Power off the server and remove the power cable from the wall jack.
2. 2.      Refer to item number 17 on figure System board components, on page 69 of the above mentioned guide to locate the system maintenance switch on the system board.
3. 3.      Set switch S6 to On position from the default Off position and the power on the server.
4. 4.      When the POST message indicates that a system maintenance switch has been turned and that it needs to be turned off, then it indicates that the NVRAM has been cleared.
5. 5.      Then power off the server, revert the switches to Off position and then power on the server.

If the issue persists, then swap with a known working HP PCIe card to isolate between the card and the slot.

Regards,

Vijayasarathy

[mattm591]

A couple of hours ago, completely out of the blue, it shup up!

It's now humming happily away at a much more normal level.

Fingers crossed it stays happy. No idea what was wrong before, and lets hope I never have to reboot and it all starts up again!

[TVVJ]

Login into the server’s iLO, if it has been configured, to review iLO event logs and IML logs for any hardware issues captured.

Also, review the operating system logs for any thermal issues.

Regards,

Vijayasarathy

[next_thomas]

Hi,
We are currently testing a ML310e Gen8 v2 server.

Same problem here : the fan goes crazy after a few moments.
It seems the problem happens only when the storage controller B120i is in used. When booting up, it reports some disks are overheating, which is not the case.
We disabled the smartarray controller and configured as SATA AHCI only, and the problem disappears.
Of course, we'd like to configure a real array using the controller.
The iLO has been updated to the latest version (1.32) and BIOS is up to date.

Any idea?
Thanks,
Thomas

[TVVJ]

Check if the firmware of the HP Dynamic Smart Array B120i Controller is the latest. You may use the Service Pack For ProLiant (SPP) to make an offline update of the controller and the hard drives in the server.

Click here to access the download page.

If the issue persists, then reseat the hard drives. If the issue persists, then ensure that only HP branded hard drives are installed.

If the issue persists, then contact HP Support with the extract of Active Health Logs for further support.

Regards,

Vijayasarathy

[mhorse_13]

I have the ML310e Gen8 updated to ROM version J04 dated 8/24/2013. iLo version 1.3 and P420 firmware to 4.68. The updated was done via SPP dated September 2013. The fan noise still there and goes up to 100% after powering up the system running ESXi 5.5 - the new firmware help a bit as the fan took longer to reach 100%.

I moved the server to an environment with 0C and fan speed were initially running at 19% Inet and 13% exhaust.

However, the exhaust fan speed goes up slowly to reach 50% when the temperate of the sensor "05-HD MAX" goes up from 44C to 47C. To stop the temperature rise, I have to put the ESX into maintenance mode.

 iLo shows this sensor location is "system"  ESXi configuration/health status says this is the Disk or Disk Bay 1 05-HD Max -- Normal when it reports the temperature of 46C

Does anybody know where the physical location of this sensor? And is there anyway to disable this sensor? Is there any HP documentation showing the physical location of these sensors?