So, long story short, I was screwing with my home server and shorted something on the board when installing video cards. The rear of the card contacted the board sort of near the 24-pin connector and the system wouldn’t turn on. Forgot to unplug from the wall, stupid mistake.
So I bought a new board (Asus Z9PE-D16/2L) and it wouldn’t boot with both CPUs and all eight sticks of RAM installed. After some (read: a lot of) troubleshooting, I have determined that the system will only boot with one stick of RAM installed for each CPU. Any more and I get a Q-Code of b7. I checked all sticks individually and they all boot fine. Q-Code b7 is “Progress NVRAM configuration reset.”
Is it safe to assume I borked one or both CPUs so they can’t run with more than one stick of RAM? They are both Xeon E5-2670.
Did you allready check if there is a newer bios available for the board?
Another thing you could try is to clear the cmos by unplugging the board from wall power, and take out the bios batteries for a few minutes.
Then put them back and check if it works and detects all memory sticks then.
Thanks for the reply.
Yes, I installed the newest version of the BIOS. I also tried clearing the CMOS and removing power. No change.
I can put either CPU in socket 1 with all eight sticks of RAM (also with socket 1) it works fine. As soon as I put a CPU in socket 2 and split the RAM between the two CPUs it won’t boot. I have double checked that the RAM is in the correct slots, too.
Have you run a MEM86 test with all RAM installed?
With one CPU and all RAM? Nope. Last time I did that (with the original board and both CPUs) it took several days to do a full memtest with 64GB RAM.
I could probably do half at a time and see what happens. Memory errors usually show up pretty quick. I am using my memtest USB on another system right now, so I can try that tomorrow.
Are you following the proper memory order when installing it?
Yes, memory is in the correct slots according to the manual.
I’ve had a motherboard die, and when replaced I found out the CPU also died, it had one black burnt dot on one of the “pins”
did some googling
this guy had the same error code witht he same board and the problem was a short. is it possible you’re board is touching part of the case? specifically around cpu 2 or inbetween cpu 2 and the ram. that motherboard is really a e atx and probably is a different size then your previous board and might not be compatible with your case.
@fredrich_nietze I just removed the board and checked all the stand offs. They are all in the correct place, in fact one was missing. The previous board was SSI-EEB, and the new Asus board appears to be the same. All the stand offs line up. This system is housed in a Rosewill 4u case. I didn’t have all the screws installed on previous boot attempts, but I just tried it with them all thinking it was a grounding issue, but it didn’t help.
@mirrorneuron I removed both CPUs and checked the pads. They all look fine, no black spots. I cleaned the pads with 91% isopropyl alcohol just because. Didn’t change anything.
I just tried one CPU in socket 1 with four sticks of RAM. I got the same b7 code. I tired the other CPU in socket 1, and got the same same code. So that’s different. Previously I could boot with one CPU and four or eight sticks of RAM. Some of the pins in the socket look funny, so I am going to remove everything again and take a close look at that. I am also suspecting the power supply a bit, so I will take the one out of my desktop and try that.
EDIT: Well, I currently have both CPUs and four sticks (two per socket) booted and running. If I put the four other sticks in I get a new code of bA. The power supply seems to be OK, as it worked with the one from my desktop and the original one.
So I am leaning towards something wrong with one (or more) of the remaining four sticks of RAM. I will go through them one at a time and see if I can find the bad one.
EDIT 2: Well, all four remaining sticks work when checked individually by putting each one in the system one at a time with the other three known working sticks. So I then replaced the four known working sticks with the other four, and lo it booted fine.
Something is preventing the system from booting when either eight sticks or 64GB of RAM is installed. I have no idea.
Yeah, memtest is probably the next step. I’ll set that up tonight and let it go.
I went through every memory setting in the BIOS, changing one at a time and checking with the remaining sticks. Nothing got it to boot with all eight sticks.
Ran Memtest86+ on the first set of four sticks of RAM, then the second set of four. Both times it passed. Went fast because I could tell it to use all the cores.
What are the chances now that something’s wrong with the CPUs? I have Asus support involved and I’m at the point where I need to tell them it passed Memtest. I don’t know if they would offer a replacement as this board is old, but in case they do I would hate to get it and have the same problem.
Edit: I’m going to do a more thorough test with Memtest86 (non +) and see what happens. Just to be sure.
Edit2: All eight sticks of RAM passed 4 passes of memtest86, running four at a time. So now I’m back to CPUs or motherboard. Again, Q-Code of bA. According to the manual, Q-Code BA is memory init.
I have no idea why this would make a difference, but it did.
Last night I decided to try getting this board to work again. I put both CPUs in with two sticks per like I know works. And it did. I let it run for a few minutes in the BIOS. Then I turned it off with the power supply switch, didn’t unplug it, and added two sticks to CPU1, but none to CPU2. Booted, and amazingly 48GB showed up. Let it sit in BIOS for a few minutes, then turned off with PS switch. Again, didn’t unplug. Added final two sticks to CPU2. Booted and 64GB showed up.
At this point, I’m not questioning it. It’s weird, but it works.
Just wanted to say that this occured for me with the Z9PA-D8. The system is working, I accidentally unseated a RAM module and didn’t push it in all the way and this same thing occured. This means this b7 error can be caused by RAM modules not being seated all the way properly.