Complete Motherboard Failure - Supermicro X9SRL-F

[NFSBuff]

Greetings all.

I am faced with a rather dire problem. Last night, I powered down my machine in order to change out the UPS. The powering down went just fine, the machine shut down as expected. I went to power it back on....

The sequence for this boot, and all subsequent boots is this:
Power switch on -> monitor blinks black -> 2 beeps -> the above screen.

No Supermicro logo, POST, no response to keyboard, no BIOS access, nada.

The machine specs:

• Supermicro X9SRL-F retail from Newegg
• 32GB of Crucial ECC RAM specific modules from the approved list of
• Intel Xeon E5-2630 v2
  
• Kingston M.2 2280 120GB SATA III boot drive connected to the PCU SATA0 header
  
• Lsi 9207-8e flashed to P20.00.00.4, IT mode
  
• 4 x 4TB WD Enterprise Drives ZFS Storage pool
• Seasonic SS-400L1U powersupply

I finished building this machine 3 weeks ago, and have gotten through all the testing but hadn't yet started using it for live data.

For the build proper, a grounded anti-static mat and wrist strap were used for the physical assembly of the machine. Prior to installing FreeNAS, the memory was burned in for 24 hours using memtest and reported zero errors. The CPU was stressed for 24 hours and maintained 41c reporting no errors. I ran jgreco's solnet-array-test for 72 hours without issue, and the S.M.A.R.T tests reported no problems. FreeNAS was installed with no plugins added, and one CIFS share active. Everything looked good to move this onto the home network for use. Then last night happened.

Things attempted so far

• Removed all PCIe cards from the system - no change
• Used one stick of RAM at a time in different slots - no change
• Reset the CMOS - no change
• Tried Supermicro's CTRL + HOME while booting with usb stick loaded with BIOS ROM - no change

Of note, the IPMI heartbeat LED on the motherboard is no longer blinking since the above image showed up.

I'm passed the 30 day exchange window with Newegg, so it'd have to be an RMA with Supermicro if it comes to that.

So what have I missed? Any suggestions are welcome.

 

[Mirfster]

Disconnect everything and only leave the basics: CPU, PSU, RAM.
Make sure everything is properly seated; re-seat to be sure.
Check to see if there is a jumper to disable BMC/IPMI; if this board has it. I have seen some boards where it would act like this when BMC/IPMI went haywire.
Just try to get it to POST and then work up from there.

 

[NFSBuff]

Thanks for the reply Mirfster.

I've attempted the re-seating a couple times now with no changes. (Taking the usual precautions of care when handling electronics). To be completely thorough, last night I busted out the trusty volt meter and measured each pin out on the PSU. Just in case there was an under/over voltage issue that might've fried something. Everything read out to spec within the margin of error of the meter. I also went ahead and tried the IPMI jumper disable as suggested and it did have an affect, albeit not a very big one. It changed the pattern on the screen from the one I first posted to a kind of red/pink/white alternating vibe:

I called up Supermicro tech support and described the issue along with my testing steps. Took the tech (who was very knowledgeable and pleasant to converse with) all of 1 minute to conclude the board had suffered a critical failure. I had an RMA number not a couple hours later. Luckily this system was still in the testing phase, so I didn't lose any data.

So I suppose that's it for this tale. The take away for me is, test till it fails. In lieu of that, keep your data backed up.

Oh, and @cyberjock, add more RAM. ;)

 

[Mirfster]

Glad to hear that SuperMicro RMA'd this for you.

 

[cyberjock]

You clearly needed more RAM. Upgrade to 4TB and see if that fixes it!