Anyone had issues with Xeon D-1520, specific board ASRock D1520D4i?

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

frogtech

Well-Known Member
Jan 4, 2016
1,482
272
83
35
I got the ASRock D1520D4I off a member on reddit and so far it hasn't been playing very nice, was wondering if anyone's had similar experience.

I was having trouble getting Windows installed, so much trouble in fact that I made Server 2012 and Server 2016 bootable USBs 2 different ways and tried them both. I started to think I was going crazy and decided to try installing onto a M620 blade...which was successful no problem...

When I try to install an OS, the Windows installer either freezes, reboots, or BSOD's DURING the installation, so typically on "Windows is copying files..." or "Getting files ready for installation".

BSODs typically range from CLOCK_WATCHDOG_TIMEOUT, WHEA_UNCORRECTABLE_ERROR, DPC_WATCHDOG_VIOLATION

It's interesting because I wouldn't think Server 2016 has outdated drivers for this platform.

I'm using non-ECC RAM which seems to work fine, I pulled it from my personal workstation. However I swapped that with the other RAM that was left in my workstation and I was able to get a bit further. I was finally able to install Server 2016, however, during a period of inactivity the machine just randomly BSOD'd with 0x124 bug check code. Some random hardware error. There's nothing installed in the machine other than the on-board mini SAS cable for the boot drive. Any reason to believe that ECC could be necessary even though the board supports non-ECC RAM?

One thing I thought was interesting is the presence of 8 pin CPU power on this board. I have another ASRock Xeon E3 v5/v6 board that only has 4 pin CPU + auxiliary molex power, and it's technically a higher power platform. The Xeon D-1520 board works fine with only 4 pin plugged in but I am wondering if not having the 8 pin connected via a 4 pin + molex adapter is causing instability?

Anyway, anyone have any other thoughts or similar experiences with this platform? Or any suggestions? Right now I am running Intel Burn Test with 10 passes set to stress level maximum (this just uses the highest available amount of RAM). I am open to other suggestions that might help determine which hardware is causing a problem.

link to board: King Fung Electronics | D1520D4I
 
Last edited:

dicecca112

Active Member
Feb 10, 2016
302
36
28
39
Did you get any memory dumps (usually in C:/Windows)? A program like windbg can read the Memory dump and give an indication of what driver was crashing. I'd be happy to help if you upload the memory dump somewhere I can access
 

EffrafaxOfWug

Radioactive Member
Feb 12, 2015
1,394
511
113
Sounds obvious, but have you made double-dog sure that the RAM is correctly seated and that the DIMM slots are free of dust bunnies? I had a board once that'd reboot randomly like this (the big advantage of ECC and the IPMI log being that it'd tell you it was a memory error), eventually turned out that despite all the DIMM clips being secure, one of the DIMMs wasn't fully inserted so whenever a particular region of memory was accessed (which might not happen for hours), the machine would log an uncorrectable memory error and reboot.

Is there anything in the watchdog logs in the IPMI interface?

A quick way to verify this in your situation would likely involve a few cycles through with a bootable memtest86+ stick.
 

frogtech

Well-Known Member
Jan 4, 2016
1,482
272
83
35
Did you get any memory dumps (usually in C:/Windows)? A program like windbg can read the Memory dump and give an indication of what driver was crashing. I'd be happy to help if you upload the memory dump somewhere I can access
Yeah I've uploaded it here:

MEMORY.DMP
 

cesmith9999

Well-Known Member
Mar 26, 2013
1,417
468
83
when you have different Bug Check errors, like your list from above. it is typically a hardware issue. I would start with replacing the RAM. then I would normally suggest replacing the CPU then MB... but in this case...

Chris
 

dicecca112

Active Member
Feb 10, 2016
302
36
28
39
I agree, DPC_WATCHDOG_VIOLATION (133) from the memory dump typically is hardware, starting with Memory in my experience. It points to ntkernal.exe, which is what typically affected when the Memory is failing.
 

frogtech

Well-Known Member
Jan 4, 2016
1,482
272
83
35
I started memtest, I don't really think it's a RAM problem. I think there is some board level issue that probably wasn't disclosed or isn't present in FreeNAS which is what the previous owner ran.
 

frogtech

Well-Known Member
Jan 4, 2016
1,482
272
83
35
I corresponded with the member to send the board back for a refund, he was nice enough to offer it. After 10 hours 36 minutes of memtest86+ there were no errors, I do not think it needs to be run for longer as this RAM has been in my personal workstation for over 2 years now without any problems. Very odd indeed.