Random Bluescreens

[Kevin]

I'm on my third reformat and can't shake these damn bluescreens. Originally I thought memory, ran MemTest for a few hours and no failures. Then I ran a Drive Fitness test on my HDD and nothing. I don't think its video because I've used both NVIDIA and ATI cards and still get the blue screens.

I can go a week without a blue screen, or I can get 3 in 1 day. Its getting very annoying and I'm running out of options.

Specs:

A7N8X-E w/ WiFi
AMD 2500+ Barton
2x 512MB PC2700 Crucial DDR
120GB Hitachi SATA HDD
Radeon 9800PRO
Lite-On LTR-52327S
NEC ND-3520A
JVC Lite-On XJ-HD165H (DVD-ROM)
Canon i560
Dell PS/2 Keyboard
MX700 Mouse
PS2 -> USB Adaptor

Running Windows XP SP2

That is basically everything connected to the computer. I have about 13 MiniDump files if anyone can use them to help me.

Thanks

 

[TGS]

Does the BSoD reference any particular device driver? Can you swap out for another motherboard?

 

[Kevin]

Sometimes it does, sometimes it doesn't. I've gotten the blue screens with 2 motherboards although they are similar, an A7N8X Deluxe, and the A7N8X-E...

 

[QueBert]

I'm in a similar boat. I get BSOD's that result in a reboot with the "windows has recovered from a serious error" message box. I have replaced EVERYTHING in my PC, and I cannot figure out why I still get them. reinstall, formats, new PSU's, new mb, new CPU. No overclocking, nothing fancy hardware wise. My PC hates me, that is what I've come to realize.

 

[Eltano1]

I have the same case with an EPOX mobo, after swapping memory, video card, NIC and finally the PS, I swapped the mobo and no more BSOD. Also I realized that some capacitors on the mobo were bulging and leaking so I eamiled EPOX and they fixed for $17.00 including shipping.

Eltano

 

[Kevin]

It happened on both the old A7N8X and now the new one. I have this gut feeling its XP SP2 causing all the problems. Usually its HDD or RAM related, but Memtest shows no errors, and the Drive Fitness Test for the HDD came back with no errors after the advanced test. I've done OS repairs, reformat and I can't figure it out. They are completely random like someone is literally rolling dice and deciding when the kill the session. I just wish I knew how to read/understand/view the mini-dumps as I'm sure some kind of answer is hidden in them...

 

[Phoenix86]

Disable the "automatically restart" after system failure (system properties, advanced, settings under startup and recovery). This will stop the PC from rebooting after BSODs, so you can tell you had one instead of a brown/black out, and actually see the message.

What is are the details of the BSOD, like what is the STOP message (IT_WILL_LOOK_LIKE_THIS) and the 0x0000000 numbers? Also is there a file mentioned?

It's not likely caused by SP2, there are little bugs in it, however it may be related. Many drivers do not work on SP2, so it could be a bad driver. Kinda grasping at straws w/o the actual error message though.

 

[dclive]

Originally posted by: Phoenix86
Disable the "automatically restart" after system failure (system properties, advanced, settings under startup and recovery). This will stop the PC from rebooting after BSODs, so you can tell you had one instead of a brown/black out, and actually see the message.

What is are the details of the BSOD, like what is the STOP message (IT_WILL_LOOK_LIKE_THIS) and the 0x0000000 numbers? Also is there a file mentioned?

It's not likely caused by SP2, there are little bugs in it, however it may be related. Many drivers do not work on SP2, so it could be a bad driver. Kinda grasping at straws w/o the actual error message though.

If anyone wants someone to look at the minidumps, I'd be happy to do so.

The chances of these errors being caused purely by SP2 approach zero. The usual reason is either bad hardware or a bad driver.

If you want me to look at the dumps, here's what you need to do first:

1. Update all drivers on your system. That's _everything_ - audio, video, chipset, mice, _everything_!!!
2. Do ALL Microsoft Windows updates. ALL of them! And if it tells you to put a driver on there, and if that driver's newer than what you have already, install it!
3. If the problem (BSOD) still happens, please run MPS Reports (see my .sig for the URL) on the machine, and send me the resulting computername.cab file via e-mail.
4. Send me the minidumps (found in c:\windows\minidump) that occurred *after* you did all the driver & Windows updates.

Thanks...

 

[Kevin]

Originally posted by: Phoenix86
Disable the "automatically restart" after system failure (system properties, advanced, settings under startup and recovery). This will stop the PC from rebooting after BSODs, so you can tell you had one instead of a brown/black out, and actually see the message.

What is are the details of the BSOD, like what is the STOP message (IT_WILL_LOOK_LIKE_THIS) and the 0x0000000 numbers? Also is there a file mentioned?

It's not likely caused by SP2, there are little bugs in it, however it may be related. Many drivers do not work on SP2, so it could be a bad driver. Kinda grasping at straws w/o the actual error message though.

Before my last reformat, I used to see the full blue screen and I'd have to wait for it to do the system dump. Now, its real quick. They are normally:
Page Fault in Non Paged Area
IRQL Not Less or Equal

Which made me think memory issues, but I've ran MemTest numerous times, the most recent being yesterday for about 5 hours and not a single error.

Also, all drivers are the newest I can find and I visit Windows Update regularly.

 

[dclive]

Great! Then I'll wait for your MPSReports and memory dumps, and we'll see if we can isolate the problem.

 

[Kevin]

Sent it all a few minutes ago, thanks!

 

[dclive]

Originally posted by: Kevin
Sent it all a few minutes ago, thanks!

Crash #1 was related to mouhid - the usb/etc driver
Crash #2 was related to ATI drivers.
Crash #3 was related to memory corruption.
Crash #4 was related to the same
The most recent crash was memory corruption.
The second most recent crash pointed to ntoskrnl.exe, which ... isn't the culprit.

You've probably got a hardware issue. Return the PC to your vendor, and have him replace it with one that works. Sorry; that's about all I've got on these dumps. They're fairly random, they happen even after reinstalls..... that's typically hardware.

Another thing to try is just a plain-jane install of XP, with no additional third party drivers. If the same thing happens, then you know it's hardware.

 

[mechBgon]

You've probably got a hardware issue.

Aha, that's my cue

1) What brand & model of power supply?

2) What memory voltage are you using, or did you leave it on AUTO

3) Are there any PCI cards installed?

4) Is the motherboard's BIOS the latest version?

 

[Kevin]

"Return the PC to your vendor, and have him replace it with one that works."

Unfortunately that would be me.

1) Antec SmartPower 450
2) Memory Voltage is Auto
3) Nope, just an Asus WiFi card in the Asus socket (bluescreens predate it though)
4) Latest I see on Asus.com

 

[mechBgon]

Ok, good. Bring your memory voltage up to 2.7 volts in the JumperFree section of the BIOS and see what effect that has over the course of time. Memtest86 was a good start, but it doesn't simultaneously load up other parts of your system like real-life usage does, so give this a try.

If you've been wanting an excuse to buy a good UPS with automatic voltage regulation too, I'll write you a note

 

[Phoenix86]

The simple fact you reloaded 3 times and still have BSODs indicates it's hardware, I'd ignore those driver error BSODs. It's likely showing a symptom, not the cause.

How many sticks of RAM? Can you test one at a time?

About voltages, what is your mobo reporting the rails at? Is the system otherwise loaded for high power draw (like 10 HDDs or 50 case fans)?

Bah!!! Who needs an excuse for more hardware.

 

[Kevin]

"The simple fact you reloaded 3 times and still have BSODs indicates it's hardware, I'd ignore those driver error BSODs. It's likely showing a symptom, not the cause."

True, but I don't know how to isolate the problem.

"If you've been wanting an excuse to buy a good UPS with automatic voltage regulation too, I'll write you a note"

I have a UPS but yes, I've been wanting to get a better one.

"How many sticks of RAM? Can you test one at a time?"

2 sticks of 512MB DDR333 in Dual Channel

"About voltages, what is your mobo reporting the rails at? Is the system otherwise loaded for high power draw (like 10 HDDs or 50 case fans)? "

I'll check the voltages when I get home in an hour or so. I don't think I have high power draw, 1 HHD, 4 case fans (3 connected to a Nexus fan bus), DVD-ROM, DVD-RW, CD-RW, Video Card

 

[Phoenix86]

Well, if you want to get to nuts and bolts fast, remove everything not needed from the system. IE, take out the optical drives, PCI cards to both remove load and possible conflicts. It's likely your problem isn't related to those parts, but just take them out of the equation.

How long are you letting memtest run? Are you getting 0 errors?

 

[Kevin]

OK, here's what Asus Probe has to say:

+12v = 12.16
+5v = 4.88
+3.3v = 3.2
vcore = 1.68

I'm guessing the 5v rail is way underpowered?

 

[mechBgon]

1) Asus probe is known to smoke crack. I would use an actual voltmeter if you have one.

2) 4.88 volts on the 5V line is under 3% off spec, that's great (if Asus probe is not smokin' crack).

3) did you try raising your DDR voltage a bit, that may help you.

Bah!!! Who needs an excuse for more hardware.

Now this guy speaks my language Let's see... the symptoms indicate a definite need for... *thinks* ... some brand-new 15k SCSI drives. Definitely!

 

[Kevin]

Well, the 5v rail seems to be varied between 4.811, 4.838, 4.866 and 4.88X. I just upped the voltage in the BIOS to 2.7 for the memory but I won't know till some random amount of time since these bluescreens really have no rhyme or reason. I also disabled Logitch Mouseware from running at startup, which sucks because I lose functionality of my zoom buttons, but hey, maybe it will solve my problem!

"Now this guy speaks my language Let's see... the symptoms indicate a definite need for... *thinks* ... some brand-new 15k SCSI drives. Definitely! "

You know, deep down I think its the SATA HDD but I can't prove it. Back in December I did a major reformat and replaced 2 WD IDE with 1 SATA. Ever since that reformat, I'd get random blue screens. My next change was motherboards, A7N8X Deluxe to A7N8X-E, still got the blue screens. I did 2 Windows XP Repairs and nothing. Finally I got sick of it all and reformatted again maybe a month ago. That time I replaced my Ti4200 with a 9800PRO. BAM, blue screens back. I was hoping it was the video card as I would get a lof of blue screens that listed nvidia drivers. But alas, I've seen 1 or 2 ATI drivers listed. I've ran Hitachi drive fitness test on the SATA 3 times and all 3 times have returned 0 errors. I'm really stumped, I have no idea what it can be. All drivers are fairly new and all are WHQL too I believe.

 

[mechBgon]

That's a reasonable suspicion. Asus had to release BIOS updates for the A7N8X Deluxe to fix SATA corruption issues with Ratpros.

I had an A7N8X Deluxe that I worked pretty darn hard, running 1.5GB of RAM as a combo fileserver/workstation down at the office, and it never missed a beat (unless I starved the RAM for voltage by using AUTO voltage, which resulted in... oh yeah, occasional BSODs! ). Then again, I never used mine with SATA. Dual 15k SCSI drives, sure.

 

[Phoenix86]

Check the drive with the manuf.'s test. You can d/l all the tools you need on the Ultimate Boot CD. It's loaded with other tools, or you can get the WD specific tools here.

At least that'll test the drive itself.

 

[Kevin]

Its not a WD, its a Hitachi. And I've already done their test (Drive Fitness Test) 3 times, with 100% success each time.

 

[Phoenix86]

Whoops... read raptor and went down that path...

You never answered, how long are you letting memtest run?