BSOD - can't figure out why

is it fully up to date with all patches?

also, worth trying is checking whether the RAM chips are ok. try taking them out and test by putting one by one.

Hi innocentdevil,

all up to date, RAM is ok.

Thx

it lists only this :  X64_0x124_GenuineIntel_PROCESSOR_BUS_PRV
so - can you :
-upgrade the bios
-test with another OS, eg 32-bit?
-another cpu ?

you can also disconnect all extra devices to eliminate possible trouble causes

Hi Nobus,
I was afraid you would make me work.
Bios is up to date. No overheating happening (it sometimes happens in the early morning).
If it is hardware - any other trick to narrow it down?

how often is once in a while?
if frequent  -- you can boot from a live cd, and check if it happens; i recommend Knoppix :
ftp://ftp.gwdg.de/pub/linux/knoppix/KNOPPIX_V6.7.0CD-2011-08-01-EN.iso

otherwise, install an OS on another disk and test with that (as said)
you can even use UBuntu

Twice in the last ten days.

It's the machine I am working on a 12hrs a day...
It crashed twice within the last ten days.
I rather try something else first - gotta earn my money.

>>  I rather try something else first   <<  good idea, but what?
i see it as the only alternative
you can install the same OS on it - it has a grace period of about 30 days

I will remove hardware, step by step.

Today it shut down - just as if the power went off. No BSOD, just poof, gone.

i agree that eliminating hardware is a good way too
disconnect extra disk and cd drives - or all cd drives
remove add-on controllers, disable sound in the bios

I was alwasy a bit suspicious about this combination case/motherboard: Sometimes just by inserting a USB stick or touching (=grounding) the USB port, it shut down. This case has also a hot-swap HDD slot which I will disable now.

>>  inserting a USB stick or touching (=grounding) the USB port, it shut down  <<

this looks like you've got static electricity around - take anti ESD measures
is the box well grounded?

never felt it when touching the box?

Once in a while when inserting a USB stick there was something like a split second discharge.
What do I ground and how?

you need to create a conductor from the pc chassis to earth, so all charges discharge to earth
so :
-check that the chassis connects to the earth prong of the AC cable - this can be measured with an ohmmeter - cost +/- 10$
-check that the AC wiring is properly earthed - or have an electrician do it for you

also - before inserting anything into USB -  first touch the case with your hand , preferable to a screw on the chassis
you can use esd mats also http://www.esdmat.com/

more info :  http://en.wikipedia.org/wiki/Earthing_system

Hi Nobus,

you are referring to grounding the pc chassis to earth.  How about static discharge motherboard - chassis as a reason for the crashes?
And how would I measure and fix that?

that is also a good reason - if you suspect that
did you ever experience it?
you don't measure it, just fix it by starting with the good ground i posted above, use ESD mat,
spray water in the air with a plant spray (it builds up mostly in very dry air) or an air moistener

I had shutdowns in two ways - with plugging in a USB stick, and without, just even by doing nothing.
If it shuts down, it does that as if you pull the plug.
It reminds me of the lights going off due to an electric shortcut...

that's no sign of ESD; normally you expect sparks, or tingles to be felt when touching the case
did you check the ground connection yet?
THAT is the first step to take

Hi Nobus,

I am out of town today, will try and let you know as soon as I am back.
Thanks!

no problem - post your findings when back

Back on the block.
All earthing is fine.

Last night, I returned to the machine, which had run all day, just moved the mouse, did not touch anything else, and:
Blue Screen with a message of Memory Mismanagement. No minidump.
I have 2 sets of 4GB (2 each) memory chips in there from different manufacturers, will exchange them one set, then the other.

Thx

a strange issue
did you touch the case BEFORE touching anything else
how did you know earthing is ok?

measured with a voltmeter as you described.

ah - ok, so you got  a 0 ohm from case to rail ground...fine then
and you touched the case first?   try that in the future

still zero.
But the variety of issues reminds me of a Dell laptop that a friend once had. Only a factory reset cured it.
Well, before I'll play with hardware.

And today another BSOD (not just a shut down), with a message about resetting the video card settings, some thing about ati***.sys.
The strange thing: I see it creating a minidump, but the folder C:\Windows\Minidump is empty!
Same thing happened with the last BSOD, I see on the blue screen the data being saved, but the minidump folder is empty.
I now deleted the folder and recreated it, since it was locked and read-only.

>>  I see it creating a minidump, but the folder C:\Windows\Minidump is empty! <<  it looks like the OS is corrupted, run sfc, or a repair install :

  http://www.sevenforums.com/tutorials/681-startup-repair.html            repair windows 7
http://www.sevenforums.com/tutorials/1538-sfc-scannow-command-system-file-checker.html         SFC

and verify that it is set for making dumps : http://blog.nirsoft.net/2010/07/27/how-to-configure-windows-to-create-minidump-files-on-bsod/

Interesting. It was set to the Kernel memory dump.
That produced several dump files, one with 1527kb, five others with 299kb, all with

BugCheck 117, {fffffa801e4904e0, fffff88004a1e768, 0, 0}

Unable to load image atikmpag.sys, Win32 error 0n2
*** WARNING: Unable to verify timestamp for atikmpag.sys
*** ERROR: Module load completed but symbols could not be loaded for atikmpag.sys
Probably caused by : atikmpag.sys ( atikmpag+8768 )

(interesting - I am on 64bit)
I already had the latest ATI driver.
Will uninstall, use Driver sweeper and reinstall.

you can also try an older driver - in several cases, the newest one  was not the best...

Oh, man. Big mess. Several programs don't work anymore, will most likely have to do a repair install. Guess Driver Sweeper swept more than supposed to do.

i never advise in favor of "automated" programs
if you have a system restore point, maybe that's the way to get it back "as was"

Imagine this: no older restore points than this morning!
Win Explorer hangs up. Many settings of programs have just disappeared. The whole Adobe Suite - no way to start it.

Win Explorer can't get to DVD, so no repair installation.
Read an error message about Visual Basic ++ and msvcrt.dll making a wrong request.
Saving now my data on C drive (most of it is on E:), one more other try, then reinstall.

does driver sweeper have a backup to get back?  if so use that

Tried that already without success. Running the machine now with the cloned drive from two weeks ago. The data corruption on the original drive is a mystery to me. It is on both partitions (a WD600 Raptor)

what data corruption?  not sure you mentioned that ?

Some images in the Thunderbird inbox cannot be displayed anymore, some psd files cannot be opened  anymore.
Some messages in  the inbox contain carbage code. Some downloaded zip files of programs that contain .exe files, cannot be installed.
I think I have to start from scratch with a differerent HDD, MoBo and CPU.

you can always test if the hardware runs from a live Knoppix CD - if ok, the hard is fine :
ftp://ftp.gwdg.de/pub/linux/knoppix/KNOPPIX_V6.7.0CD-2011-08-01-EN.iso

I am pretty sure now that the hard drive - a WD600 Raptor was the cause.
Especially the disaster on Thursday - lots of files corrupted, restore points disappeared.
I had the drive cloned two weeks ago and the machine runs from a Seagate drive without major issues.
Since IO could not repair the OS, I installed another copy of Win7 on the OS partition and got some error message when I restarted.

so all is ok now?

So far so good.
Have to see. What really annoys me is the file corruption in my emails.
Attachments images are unreadable, emails contain spaghetti code.

that could also be caused by the raptor
maybe you can repair it with hddregenerator, it fixed many drives for me :
http://www.dposoft.net/hdd.html

So far no luck with anything. Anyway - that another issue.
I'll assign points soon.

Going to another hdd did the trick.
Sadly, some of my data remained unretrievable, although the Raptor showed no errors, all sectors fine.
My guess is that it is an electronic issue in the drive.

Thanks!

can be - you can always run a diag on it, to know it's status :
http://www.ultimatebootcd.com/                              ultimate boot cd
http://ubcd.mirror.fusa.be/ubcd511.iso                        direct link ultimate boot cd