Apologies in advance, I'll be the first to admit that this board is way more than I need and probably beyond my tech level, but I'm not trying to do anything fancy like overclocking here (blasphemous, I know, but moving on...)
 
Processors: 2x Xeon Nehalem 5520
Memory: 6x 4gb ECC DDR3-1066 quad-rank in dimm slots 1,3,6 of both banks
 
BIOS and various OS's all recognize both CPUs, but only 12gb of memory is recognized, and by pulling and swapping modules around I've figured out it's the second CPU's dimm slots that aren't recognizing anything in them.
The only modification I've made to bios settings is changing sata from ide to ahci and altering the boot order.
 
I can't believe the board's faulty to the point of six dimm slots not working without exhibiting other symptoms
 
What am I doing wrong?

Hi!
I also had 5520's before. Sorry if you already did this, I guess you've tried using the same dimms that you put in second CPU (CPU1) in CPU0?
 
I had similar problems mixing and matching 2GB and 4GB dimms, they had to be put in certain order or they wouldn't be recognized. I know all your dimms are the same. Sometimes it's like if you put a certain dimm in a certain slot, it won't work and make others don't work either.
 
Have you tried using just 1 dimm for CPU0 and 1 for CPU1?

check ram all in cpu 0( 6dim populated red an black )  disabled cpu1
in bios verify if you have 24gb
adjust voltage and timing manualy, speed ram
check in windows  you  have 24gb
put 3 dim on cpu1  fix the cpu is enabled ,
check in bios if you have 24gband 2 cpu in cpuz
 
 

1 dimm with each cpu and cpu1 disabled: 4gb and 1 processor
1 dimm with each cpu and cpu1 enabled: 4gb and 2 processors
 
6 dimms in cpu0 bank with cpu1 disabled: 24gb and 1 processor
6 dimms in cpu0 bank with cpu1 enabled: 24gb and 2 processors
3 dimms in each bank regardless of whether cpu1 is enabled/disabled: 12gb
 
manually setting the memory frequency to 1066mhz didn't change anything either :-(

"3 dimms in each bank regardless of whether cpu1 is enabled/disabled: 12gb"
3 dimms for CPU0 AND 3 dimms for CPU1?
Or just 3 dimms for CPU0 and 0 for CPU1
Or 3 dimms for CPU1 and 0 for CPU0

I've heard stories if the CPU cooler is to tight it can cause loss of memory banks.
Try backing off on the heatsink bolts a little for that CPU.
 
Edit: if in a case it could be a misplacement of a standoff causing problems.

It actually hadn't occurred to me to try dimms with just cpu1, so i just went and tried that:
 
0 dimms cpu0, 3 dimms cpu1 results in error codes via the onboard LED in order: 00, E8, 02, 69 repeating. None of those are in the book, but it's pretty obvious the board's complaining about having no ram.
 
What else can I try?

Sorry this should be the most important clue:
 
1 dimm with each cpu and cpu1 disabled: 4gb and 1 processor
1 dimm with each cpu and cpu1 enabled: 4gb and 2 processors
 
 CPU1 doesn't even see one dimm. Have you tried in different slots? To rule out compatibility, do you have other old dimms to test for CPU1?

macdamon


It actually hadn't occurred to me to try dimms with just cpu1, so i just went and tried that:

0 dimms cpu0, 3 dimms cpu1 results in error codes via the onboard LED in order: 00, E8, 02, 69 repeating. None of those are in the book, but it's pretty obvious the board's complaining about having no ram.

What else can I try?

See my other post above.
 
Edit: If in a case setup you may want to remove the motherboard and do a bench setup test out side the case.
If it stil doesn't see the memory chips on those banks, I'd say RMA time.
You could also try swapping the CPU's around to the other socket, one of the CPU's could be the problem.

Tried borrowing a couple of 2gb ddr3 modules from another system, both recognized if placed with cpu0, none recognized if placed with cpu1, only one recognized if split between the two cpu's...
 
I tried backing off the heatsink, that caused a 8°c rise in temp for cpu1 and no benefits
 
I'll go pull it out of the case now, since that's my first step to RMA it anyway... will post back in a bit about whether that had any effect

Just the step TheWolf said remains, swap the CPUs to see if the problem is with the CPU or the motherboard.

removing it from the case has (temporarily I hope) cost me all diagnostic capability... all I get now is LED codes CF,F1,F3 flashing in rapid succession... no mention of those in the manual... what've I done wrong now? Does it require that a case fan be connected or something?
 
edit: nope, nothing to do with case fans... I have a sinking feeling CF means CPU Failure... which's bad on many levels because I know those two CPUs were good, they're castoffs from an upgrade to another machine
 
edit2: my current problem seems to be addressed here:
/forums/fb.ashx?m=1245788
 
left it unplugged, will come back to it later...
 
edit3: came back, did a bios reset, all's well on the booting front again, outside the case now

Okay, outside the case now, swapped the CPUs, still nothing recognized in the cpu1 dimm slots.
 
Guess I'll be filing an RMA, unless somebody has another idea!
 
Thanks to both geort45 and TheWolf for the suggestions

RMA time it seems :(

geort45


RMA time it seems :(

About the only thing lift I can think of is doing a fresh flash of one of the bios chips.
Maybe try with a bios version you haven't tried before now.
I'd flash from within bios or threw a DOS flash floppy disk or bootable USB pin drive.
Don't use any windows flashing as that could be part of the problem, corrupt flash.
I think there is a sticky on the best way to do a flash or see the latest bios sticky for the info.

macdamon


It actually hadn't occurred to me to try dimms with just cpu1, so i just went and tried that:

0 dimms cpu0, 3 dimms cpu1 results in error codes via the onboard LED in order: 00, E8, 02, 69 repeating. None of those are in the book, but it's pretty obvious the board's complaining about having no ram.

What else can I try?

Here are several (easy to do) things you might want to do to confirm there is a hardware problem:
 
1-      Verify the actual memory voltage values. If you have a voltage meter, find the voltage pads on the motherboard and check the voltages for DIMM (memory bank 0 voltage) and DIMM_1 (memory bank 1 voltage). If a volt meter is not available, you can verify that memory voltages in the BIOS under Hardware Health Configuration.
2-      Verify that the memory voltage configurations were set correctly. You can check this in the BIOS under Voltage Configuration. Look at the values for CPU0 and 1 DIMM voltages. At the same time, look at the other voltage settings.
3-      Verify whether the system is reading the installed hardware correctly. That is, after the system boots, check the content of the system management bus (SMBUS). If operating in Windows, you can easily do this by running HWMONITOR and saving the “Monitoring data” (a file that you can download). The content of this file shows you what the motherboard is told when the operating system boots. The file lists the various chips on the motherboard and enumerates their topography along with configuration parameters and values. Scroll down to the “PROCESSORS” and verify that the information is appropriate for the two CPUs installed.