GA-X58A-UDR3 HDD write errors

[jaybird16]

My system had been working fine for about 6 months before I started to have corrupted data and BSOD issues. I looked at the logs and noticed quite a number of Event ID 51 errors (among others). I replaced my hard drive (WD Caviar Black 640 - SATA2 running ACHI off of the Intel controller) with a 1T Caviar Black which worked for a couple days until I noticed more event ID errors that dealt with the controller. I switched the cable for the HD and moved between the intel controller, the SATA2 and the Marvell, all worked for a little (between 1 day to 3 days operation) before I continued to get controller errors and eventually BSODs. I've tried IDE, ACHI modes on all controllers. I tested the memory which tested clean. Are there any other suggestion you have to troubleshoot this problem, or is this a motherboard hardware issue at this point?

-------------------------------------------------------------------------------
Model Name : GA-X58A-UD3R(rev. 2.0)
--------------------------
M/B Rev : 2.0
BIOS Ver : FA
Operation System : Win 7 64-bit      SP :
Memory: G.Skill      Type : ddr3

WD 640 Black SATA2
WD 1T Caviar Black SATA3

[Dark Mantis]

Hi and welcome to the Gigabyte Forum.

You have already done a lot of the checking but at the moment we don't seem to be that much further  forward.

When you said you checked the RAM did you use Memtest86+ or the beta version for the X58 chipset?

How many times did you run the loop before changing modules?

Did you run the modules seperately?

What are the temperatures of your system like in normal running?

Which processor do you have installed?

Sorry for all the questions but there will probably be even more to come.

[jaybird16]

Hi,

I ran the memtest86+ for 3 cycles with all 3 2G modules at once. I ran each individual module for 1 cycle only. (all clean) I don't recall the temperature, but it was not clocking something that concerned me when I looked at the readings in the BIOS. My processor is a i7-920 with stock cooling, if that matters.

thanks for the quick response!

[Dark Mantis]

Right processor first. As you are using the stock cooler check each of the four legs and make sure they are all fully, firmly inserted through the motherboard. This is a common problem and a bit to much heat build up could cause data corruption.

Run Memtest again but use the beta version downloadable from the same site for the X58 chipset. This time run on each module for at least ten loops.

Post back when that is completed please.

[jaybird16]

The stock cpu fan is securely attached. CPU temp after running the memtest for several hours is 45C. All the memory modules passed 10 iterations of the test.

[Dark Mantis]

OK so hopefully we can rule these out of the equation.  There is a mass of info regarding theis error on the web but knowone seems to be sure of what is causing it. Sometimes it seems to be softwar and others hardware. The most probable cause is a hard drive that is going to die. It could of course be the processor as the memory controllers are now a part of that and so can cause a damaged data problem. The only one easy thing to try would be changing the data cable on the drive. Cheap and easy.
Sorry I really can't be more specific than that.

[jaybird16]

Thanks for the help. I've swapped the data cables twice with new ones, early in the troubleshooting session. Given the fact that I've re-installed win 7 several times, and had this system working for 6 months prior to my difficulties, this points to a problem with one of the hardware components.

At this point, not knowing hardware architecture, would the CPU or motherboard be more likely the culprit?

[Dark Mantis]

To be realistic it could be either. If it was the slightly older hardware I would have said motherboard for sure but now that they have put the memory controller on the CPU it is quite possibly that. Actually after just reading your original post again and seeing that you had tested the memory and it was fine I would say that it is more likely to be the motherboard. I think you need to RMA it or take it back to your retailer for RMA.