Random reboots...

[SkuLLyRT]

I began experiencing this problem about a month ago, after about a year or so of stable usage out of this build. My specs are as follows -

Opty 165 @ 2.6
2GB Crucial Ballistix DDR
Asus A8N32-SLI
EVGA 7900GT
Seasonic 600W PSU
WD 320GB HD
SB X-FI
16X DVDROM
XP PRO SP2 (32bit)

Now recently my computer has started to reboot itself in the middle of playing games and watching videos. Occasionally it will hang on the Windows "Welcome!" screen (I see "Welcome!", and then the screen goes black and my LCD goes into power saving mode) after these reboots, other times it'll go straight back into Windows like nothing was wrong.

I've Orthos'd my CPU at both stock and overclocked speeds for 24+ hours and have come up with no errors. I've Memtest86'd my RAM at both stock and overclocked speeds for 24+ hours and have come up with no errors. I've run the WD Diag Extended test (which took forever btw) and come up with no errors on my HD.

Any ideas? I'd try just swapping the video card and / or PSU, but I don't really have the cash to run out and buy extra parts for the sake of testing at the moment. I'm leaning towards the vid card, but I'd hate to send it in for RMA only to find out it was something else wrong with my system.

 

[Dahak]

Try turning off the automatic restart, in the system properties -> advanced tab -> Startup and Recovery Section -> Setting button and remove the check beside automatic restart.
What this will do is give you the blue screen stop/error message and it will help narrow down the possible problem.
Also you could run a chkdsk /f /r for a check of filesystem error which casuse reboots too, even though the wd test passed

 

[oynaz]

If the PSU is only 1 year old, I doubt that is the culprit.

It is probably the RAM. Memtest is not foolproof.

 

[SkuLLyRT]

Alright, I've turned off the automatic restarts and am about to begin checkdisk.

Are there any other tests that I could run on the RAM while I wait for my next random reboot?

 

[dclive]

Sure - analyze the dumpfiles (see my website...bsod hell section) and post the results of '!analyze -v' here.

 

[SkuLLyRT]

CHKDSK ran and cleared with no problems.

Since I haven't had any crashes yet today, I analyzed the dump files from the past 3 days. The first two look pretty familiar.

FOLLOWUP_IP:
nv4_disp+15a7b0
bfb2f7b0 d88e4c010000 fmul dword ptr [esi+14Ch]
SYMBOL_NAME: nv4_disp+15a7b0
FOLLOWUP_NAME: MachineOwner
MODULE_NAME: nv4_disp
IMAGE_NAME: nv4_disp.dll
DEBUG_FLR_IMAGE_TIMESTAMP: 472005da
STACK_COMMAND: .cxr 0xffffffffb12925ac ; kb
FAILURE_BUCKET_ID: 0x7E_nv4_disp+15a7b0
BUCKET_ID: 0x7E_nv4_disp+15a7b0
Followup: MachineOwner

----------------------------

FOLLOWUP_IP:
nv4_disp+164eb1
bfb39eb1 8b542420 mov edx,dword ptr [esp+20h]
SYMBOL_STACK_INDEX: 0
SYMBOL_NAME: nv4_disp+164eb1
FOLLOWUP_NAME: MachineOwner
MODULE_NAME: nv4_disp
IMAGE_NAME: nv4_disp.dll
DEBUG_FLR_IMAGE_TIMESTAMP: 472005da
FAILURE_BUCKET_ID: 0x50_nv4_disp+164eb1
BUCKET_ID: 0x50_nv4_disp+164eb1
Followup: MachineOwner

The last one, however, had something a little different in it.

FOLLOWUP_IP:
Ntfs!NtfsPrepareBuffers+31e
ba546022 2bf7 sub esi,edi
SYMBOL_STACK_INDEX: 0
SYMBOL_NAME: Ntfs!NtfsPrepareBuffers+31e
FOLLOWUP_NAME: MachineOwner
MODULE_NAME: Ntfs
IMAGE_NAME: Ntfs.sys
DEBUG_FLR_IMAGE_TIMESTAMP: 45cc56a7
STACK_COMMAND: .cxr 0xffffffffbaceb10c ; kb
FAILURE_BUCKET_ID: 0x24_CODE_ADDRESS_MISMATCH_Ntfs!NtfsPrepareBuffers+31e
BUCKET_ID: 0x24_CODE_ADDRESS_MISMATCH_Ntfs!NtfsPrepareBuffers+31e
Followup: MachineOwner

What now?

 

[robisbell]

go to http://www.ultimatebootcd.com/ burn it to a cd, then run memtest86+ for 6 hours minimum.
report back the results.

 

[practor]

4.4.4.8 ?

 

[dclive]

The first two dumps state that the nVidia driver caused the BSOD, so ... is your nvidia driver up to date? What is the date on it?

Are you fully up to date with all Windows OS patches from update.microsoft.com ?

 

[SkuLLyRT]

Originally posted by: robisbell
go to http://www.ultimatebootcd.com/ burn it to a cd, then run memtest86+ for 6 hours minimum.
report back the results.

Just ran that for another 7 hours and it came back with no errors.

Originally posted by: practor
4.4.4.8?

I'm assuming this is a reference to my RAM timings? They're actually at 3.4.4.8 if that's what you meant (and that's how it's been for a year or so now). If that's not what you meant please elaborate.

Originally posted by: dclive
The first two dumps state that the nVidia driver caused the BSOD, so ... is your nvidia driver up to date? What is the date on it?

Are you fully up to date with all Windows OS patches from update.microsoft.com ?

Yep, everything is up to date. 169.01 beta (from late Oct I believe), 163.75 WHQL (from early Nov)... they've both caused random reboots. I let Windows Update do it's thing every week, so we're good there. Just did a virus / spyware scan, and there doesn't appear to be any major problem there either.

 

[robisbell]

okay, download and run http://www.cpuid.com/pcwizard.php check temp and voltages, post if you would be so kind, then run the benchmark and post it.

also, on UBCD, run the appropriate HDD diagnostics if you would be so kind, and post the results.

 

[dclive]

Originally posted by: SkuLLyRT

Originally posted by: robisbell
go to http://www.ultimatebootcd.com/ burn it to a cd, then run memtest86+ for 6 hours minimum.
report back the results.

Just ran that for another 7 hours and it came back with no errors.

Originally posted by: practor
4.4.4.8?

I'm assuming this is a reference to my RAM timings? They're actually at 3.4.4.8 if that's what you meant (and that's how it's been for a year or so now). If that's not what you meant please elaborate.

Originally posted by: dclive
The first two dumps state that the nVidia driver caused the BSOD, so ... is your nvidia driver up to date? What is the date on it?

Are you fully up to date with all Windows OS patches from update.microsoft.com ?

Yep, everything is up to date. 169.01 beta (from late Oct I believe), 163.75 WHQL (from early Nov)... they've both caused random reboots. I let Windows Update do it's thing every week, so we're good there. Just did a virus / spyware scan, and there doesn't appear to be any major problem there either.

The bluescreens are being caused by the nVidia driver (or related subsystem), per the debug you did.

Do you have an ATI (or non-nVidia) card you could put into your machine for a little while to confirm this?

 

[robisbell]

I think it's a hardware issue.

 

[practor]

I'm assuming this is a reference to my RAM timings? They're actually at 3.4.4.8 if that's what you meant (and that's how it's been for a year or so now). If that's not what you meant please elaborate.

Yes m8 thats right.

 

[SkuLLyRT]

A quick update.

I got tired of re-running all of these diagnostic tests that I've run once already in the past week, so I borrowed a cheap video card (nvidia 7600 or something along those lines) from a friend and tossed it in my system. I experienced absolutely no crashes yesterday. I've got more time to stress my system today, but if I get no more crashes this weekend would it be safe for me to assume that my 7900GT is bad?

 

[dclive]

Yep.

This is why analyzing dumpfiles is the first thing people should do when there's a BSOD problem on their computer.

 

[robisbell]

dclive, you were saying it was drivers, and not hardware.

skullyrt, glad that it was figured out, I was just awaiting info, but glad you got it fixed, but I was notcertain what hardware was at issue.

 

[dclive]

Originally posted by: robisbell
dclive, you were saying it was drivers, and not hardware.

skullyrt, glad that it was figured out, I was just awaiting info, but glad you got it fixed, but I was notcertain what hardware was at issue.

Please read my post from 11-30 at 11:58 for what I actually said.

If you have any questions on that please post.

Reading the dumpfiles whenever a box bluescreens is always the first thing to do. The debugger tells you where to look.

 

[SkuLLyRT]

So far everything still appears to be working well (though TF2 and COD4 are struggling on this card, lol). Just wanted to thank all of you guys for helping me out and teaching me some new things about troubleshooting.

 

[robisbell]

not a problem. just remember the golden rule I was taught by some computer tech that had been around since the 50's. always thoroughly check hardware before checking software.
Ultimate Boot CD will be the best tool you'll need to do most of the checks, it and a multimeter.

 

[dclive]

Originally posted by: SkuLLyRT
So far everything still appears to be working well (though TF2 and COD4 are struggling on this card, lol). Just wanted to thank all of you guys for helping me out and teaching me some new things about troubleshooting.

Happy to help. Keep the guide to reading the memory dumpfiles handy - it's a good way to solve these issues.

 

[robisbell]

let me know if there is anything I can do to help, skullyrt.