[sievers]

Is there a resource out there for debugging BSOD? Maybe a list of things to check or likely causes, or some way to test to narrow down what the problem is?


In my case, I built my htpc about a week ago, was fine from the very first boot, but in the last couple days, I now get random BSOD. Sometimes it will happen right away, before windows is up, sometimes it's just before/during wmc starting up, other times I'm running for x hours and then it happens. I have not changed my hardware since I originally put it together, which is as follows:


CPU INTEL|CORE I3 2100 3.1G 3M R

MEM 2Gx2|G.SKILL F3-12800CL9D-4GBXL

MB ASROCK|H67M (B3) H67 LGA1155 R

HD 2T|WD WD20EARS 64M

TV CARD CETON | INFINITV 4 RT

BLURAY BURNER LG WH12LS30

PSU SEASONIC| S12II 380B 380W RT

HTPC CASE ANTEC|FUSION REMOTE BLACK


Seems fairly standard, I used renethx's guide to help pick these.


Software is win7 home premium. I've added chrome browser, and makemkv, and that's about it.


I've been able to read the BSOD errors a few times but it seems like it's something different every time, so not sure that it's helpful. The only time I've had this type of issue before was years ago when I built a pc and was overclocking it, so then the solution was easy. I am not overclocking anything on this machine.


Here's my current ideas, please add whatever you think would help.

1. Disconnect non-essential components to determine if they are causing the issue. This would be the ceton and the blu-ray drive. (by the way, if you've read some of my other posts, I did change my cablecard in the ceton at one point but now I'm back on the original one)

2. ??? that's all I've got!


Suffice it to say, I have not returned my cable box yet!

Thanks avs

 

[Calcvictim]

try booting in safe mode, check the event viewer for logs, events relating to kernel panicks, go from there

 

[StardogChampion]

Does it give you the name of a file in the line "The problem seems to be caused by the following file: somebaddriver.sys"


If it does, figure out which driver that is and try to roll it back to the previous version. If the system has been working fine and recently became unstable my first suspect would be a driver update.

 

[Scallica]

You need to verify that your memory modules are working properly using Memtest.

http://www.memtest.org/

 

[ChairTwo]

When you get random results from memory dump files its usually bad memory. It could also be some compatibility problem with your motherboard and memory. The memory you listed is not on the supported memory list http://www.asrock.com/mb/memory.asp?Model=H67M but it shouldn't really matter. Check the bios and make sure the memory settings match what it's rated at or even under clock it a bit to see if it makes a difference.

 

[sievers]

I ran memtest and immediately 3 errors popped up. Not familiar with memtest, I hit esc to start it again, this time I got 1 error immediately. It's still running, but seems like it's safe to say this is the problem. Is there anything to be done other than exchange the memory? Maybe I'll pull it out and reseat it and run the test again, or swap the slots. Truthfully tho, seeing these errors, I don't think I want anything to do with these sticks at this point! I read somewhere that I should test each stick individually, but then further down the thread someone else said this is not necessary, so right now it's testing with both sticks in.


ChairTwo, regarding the match with the mobo, yes maybe I should get different memory. I recall a recent thread tho where the need to follow the supported memory list was called into question, and you seem to be saying the same thing. So having seen errors in memtest (8 or more just popped up in the 2nd pass), does that indicate strictly a memory issue, not related to compatability with the mobo? Or am I asking for trouble by getting the same memory again?


I must say, I was really annoyed about this but to at least see a problem identified is awesome, I feel much better! Thanks again everyone!

 

[sievers]

Looking more closely at the memory list for the mobo, I see they list F3-12800CL9D-4GBNQ, which is basically the same specs as the one I have, but without the heat spreader. So maybe they just missed it.

 

[hettar987]

Looks like you've got it.

For future reference, there is also this tool ( www.nirsoft.net/utils/blue_screen_view.html ) that will help you decipher the content of the BSOD. It helps nailing down a culprit (driver for instance).

I'm not sure it helps with a hardware problem, so running first a memory test is always a good idea.


Of course, it didn't help me much with my BSOD, as it only tells me ntoskrnl crashed, but not the underlying cause. I guess I'll swap in and out parts to figure it out...

 

[Calcvictim]



Quote:
Originally Posted by sievers /forum/post/20856378


Looking more closely at the memory list for the mobo, I see they list F3-12800CL9D-4GBNQ, which is basically the same specs as the one I have, but without the heat spreader. So maybe they just missed it.

If the specs are the same you could just have a flaky dimm, there is no point testing individual dimms since you most likely bought both as a set.


If you're going to exchange then just get what's on the tested list to save yourself the headache.




I would try reseating the memory before returning it.

 

[ChairTwo]

Any errors in memtest are bad it doesn't give out false readings. After reseating the memory I also like to test them one at a time then I test the slots. At work I occasionally get bad memory from our distributor it's annoying.

 

[sievers]

Interesting, this morning I reseated both sticks and now memtest is running without errors. I will give it a chance, one more BSOD and it's going back!