Proliant ML350 G9 Temperature sensor 40 location

[Marcuspd]

Hi all, I wonder if anyone can help?

A customer has an ML350 G9 which is less than 2 weeks out of warranty (typical). It reboots seconds after powering on and iLO recrods a critical temperature alert at sensor 40, location I/O board, with temperatures around 90 degrees C.

This server had sat unused for a few months, then moved carefully to another building and developed this fault as soon as it was powered up. The fans appear to be working normal, but I'm told one of the 4 SAS drives, once removed, now won't go back in any slot, even the one it came out of. 

As far as I'm aware, it has no PCI add in cards and has a RAID controller. Being out of warranty, I'd like to nrrow down, system board, drive cage board or RAID controller, before stabbing in the dark. 

Cheers. 

[Ash_1711]

Greetings!

                 We would suggest the below action plan after taking into consideration all the information:

1. We can check the temperature output to find the sensor device from the command hplog -t & see it points out to any particular sensor IO device or check for Internal health LED in front of the server & try with minimal configuration (1 Processor 1 DIMM) & swap the processors & check with minimal configuration.

2. But since you mentioned server reboots as soon as it's powered on we can check the current BIOS version & update to latest version available for this model from hpe site, however, the server might need to be in warranty to download the BIOS firmware HPE site.

3. The last option will be to replace the system board as best guess part as the server was unused for months & temperature sensor could be from the I/O device.

Regards,

[Marcuspd]

Thanks for the suggestions. Unfortunately it shuts down with a red health warning LED error caused by the temperatiure alert within a few seconds of starting POST.

Would it still be possible to try updating the BIOS if it won't stay on for more than 10-15 seconds from powering on?

I've tried minimal configuration, but unfortunately it only has one CPU, so I can't rule out that one, hence why I need to know the loaction or purpose of tempersature sensor 40, before I start throwing parts at it. 

I've tried it with 1 power supply powered, RAID controller, drives and drive cage unplugged and 1 RAM module at a time. It still develops a critivcal temp alert within seconds of powering on. The fans are working fine, so all we have connected is the system board, powered by one PSU, 1 RAM module and 1 CPU.  The system board looks the most obvious, but I;d like to confirm the purpose and location f that sensor 40. 

[Torsten.]

Sensor 40 is near PCI slot 3 on the system board.

When it is reporting 90 degrees, it is most likely broken.

However, you can try update ILO to 2.70 and BIOS to 2.72:

https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_193fb26e66614be5b23034ece8

Download the file, extract "P92_2.72_03_25_2019.signed.flash" from it and use this file just like you use an ILO update *.bin file.

So you can upgrade the BIOS even if the server is powered off.

What BIOS is currently installed?

[deniserd]

Hi, I have the hp proliant ml350 g9 and wanted to reduce the fan speed as I keep the PC in my room, extremely loud.

I want to do the solution, could you please guide me where is the Ambient Sensor exactly and how it looks like (a picture would be amazingly helpful)

I put 2 fans straight on to the heating sink metals and not running the other fans.

Or any other alternative solutions like replacing the fans etc?

Thank you so much for your help.

Denis

[Sunitha_Mod]

Hello @deniserd,

Thank you for posting! Since you have posted in an old topic and there is no response yet, I would recommend you to create a new topic using the "New Discussion" button so the experts can check and guide you further.