XP blue screen -- how to get actual error?

V

venm11

2[H]4U
Joined
Oct 6, 2004
Messages
2,236
One of my XP machines is crashing intermittently. When it blue screens, a blue screen with error appears for a split second but then the system restarts. Nothing appears in the event log regarding the incident -- I suspect that the event log isn't able to be written.

Is there another place to look for a dump? How can I make the blue screen wait for reboot (which I thought it did by default)?

The actual problem: some intensive tasks trigger it quite reliably (defrag, burning cds, loading games) but not others (such as benchmarking tools, copying files btw disks, etc). All temps and voltages are within limits. The suspicious thing is that CPU load is consistently 20-30% without any process obviously consuming that cpu in task manager (cpu speed isn't the issue either).
 
Last edited:
Right click computer > properties > advanced > startup and recovery > untick automatically restart.

I believe the dump file is in %windir% but not 100% sure. Should be titled minidump.
 
Thanks! That was *very* helpful.

Question --- how reliable are those memory addresses in the blue screen, considering relative addressing? I found some intermittently bad memory addresses w/memtest86+ and it would be useful to correlate those.

The error is consistently:
driver_irql_not_less_or_equa
stop: 0x000000d1 (ox00000060, 0x00000002, 0x00000000, 0xba5fcdb0)
nvatabus.sys - address BA5fcdb0 base at ba5ea000, datestamp 420d65b5

I have yet to locate that file either in the nforce (nforce4 ultra) driver set or on my c: drive. In theory I could compare the two (eg, md5sum) to see if mine is corrupted, since there have been disk errors (chicken and egg in this scenario).
 
Sounds like you have bad memory. Try removing one stick and then seeing if the error continues. Then the other. Then try both in different slots. If it is bad memory, you need to replace them.
 
hedron said:
Right click computer > properties > advanced > startup and recovery > untick automatically restart.
Click to expand...
I am truly amazed, here in 2009, that this isn't one of the first things EVERYONE does to their systems (at least if they still have XP).
 
DeaconFrost said:
I am truly amazed, here in 2009, that this isn't one of the first things EVERYONE does to their systems (at least if they still have XP).
Click to expand...

I'd have to speculate that it's because your systems are relatively unstable. Of the 12 PCs I own, blue screens are so rare that I've never had to look into it.
 
hedron said:
Sounds like you have bad memory. Try removing one stick and then seeing if the error continues. Then the other. Then try both in different slots. If it is bad memory, you need to replace them.
Click to expand...

Well, i got them as a pair (crucial ballistix) so I'd have to RMA both. I'll try pulling them.

These sticks I replaced my OCZ sticks on which the heat spreader had warped off. The problem with this mobo is that the stock amd heatsink vents heat directly onto the dimms. I placed the memory on the far 2 slots rather than the close ones but that may not be enough. There are many complaints on newegg that these dimms overheat and fail.
 
venm11 said:
Question --- how reliable are those memory addresses in the blue screen, considering relative addressing? I found some intermittently bad memory addresses w/memtest86+ and it would be useful to correlate those.
Click to expand...

It is a memory error. Who cares what the address is? Time for some new RAM...
 
hedron said:
Sounds like you have bad memory. Try removing one stick and then seeing if the error continues. Then the other. Then try both in different slots. If it is bad memory, you need to replace them.
Click to expand...

nvatabus is nVidia ATA bus. Sounds like bad drive controller - either hardware or software.
 
REDYOUCH said:
It is a memory error. Who cares what the address is? Time for some new RAM...
Click to expand...

Well... there is probably a memory error. If I can reproduce the memory error in different situations (eg, different slots, different test tools, same address) then that would confirm it. That doesn't rule out a corrupted driver or something going south with the integrated chipset. I don't know enough about how the memory and the testing tools work to rule out a false positive due to it being mapped to a device, or DMA interference, etc.

So, I'm testing sticks separately (Hedron's idea above), as well as my old OCZ memory to see if it's actually the mobo. I'm suspicious of slot 4. I'm running the sticks in slots farthest from the CPU because of the way the heatsink vents onto them, and that's always been fine with this mobo.
 
venm11 said:
I'd have to speculate that it's because your systems are relatively unstable. Of the 12 PCs I own, blue screens are so rare that I've never had to look into it.
Click to expand...
You'd be speculating incorrectly then. It's a common sense troubleshooting step to disable automatic reboots in the event of a BSoD. I couldn't tell you when the last time I saw one a BSoD on a system I set up and/or built. I'm not sure how you could twist my words into me somehow stating my systems were unstable from that. XP's been out for 8 years now or so, and from the very beginning, this is one of the very first configuration changes most would change. How else would you ever troubleshoot a BSoD if you encountered one?
 
DeaconFrost said:
You'd be speculating incorrectly then. It's a common sense troubleshooting step to disable automatic reboots in the event of a BSoD. I couldn't tell you when the last time I saw one a BSoD on a system I set up and/or built. I'm not sure how you could twist my words into me somehow stating my systems were unstable from that. XP's been out for 8 years now or so, and from the very beginning, this is one of the very first configuration changes most would change. How else would you ever troubleshoot a BSoD if you encountered one?
Click to expand...

By finding out how to change the blue screen option, perhaps with a quick post to hardforums. Sorry for "twisting your words around" but posting here simply to pontificate serves no purpose.
 
venm11 said:
Sorry for "twisting your words around" but posting here simply to pontificate serves no purpose.
Click to expand...
Neither does twisting the words of, or talking down at the people trying to help you. Anyway, it seems as though you are on the right track to the solution. I recently had a memory slot issue with my brother-in-law's PC. It failed a memtest run, but the memory was relatively new, so I swapped sticks and slots. Sure it took much longer than expected, but at least I knew the mobo was the issue, and not the RAM.
 
venm11 said:
Well... there is probably a memory error. If I can reproduce the memory error in different situations (eg, different slots, different test tools, same address) then that would confirm it. That doesn't rule out a corrupted driver or something going south with the integrated chipset. I don't know enough about how the memory and the testing tools work to rule out a false positive due to it being mapped to a device, or DMA interference, etc.

So, I'm testing sticks separately (Hedron's idea above), as well as my old OCZ memory to see if it's actually the mobo. I'm suspicious of slot 4. I'm running the sticks in slots farthest from the CPU because of the way the heatsink vents onto them, and that's always been fine with this mobo.
Click to expand...

If you get errors in MemTest86, then there are no drivers involved. This is not rocket science, just swap the memory around to isolate the bad DIMM (or DIMMs).
 
REDYOUCH said:
If you get errors in MemTest86, then there are no drivers involved.
Click to expand...
That's why I never trust a memory diagnostic tool that runs within Windows. There are too many extra variables. Using Memtest on it's own, pretty much brings you down to bad hardware as the only cause.
 
DeaconFrost said:
Neither does twisting the words of, or talking down at the people trying to help you. Anyway, it seems as though you are on the right track to the solution. I recently had a memory slot issue with my brother-in-law's PC. It failed a memtest run, but the memory was relatively new, so I swapped sticks and slots. Sure it took much longer than expected, but at least I knew the mobo was the issue, and not the RAM.
Click to expand...

Review your first post and explain how that statement could help this issue. The tips you are giving are simply not of any use, we're way beyond that. Please... enough.
 
REDYOUCH said:
If you get errors in MemTest86, then there are no drivers involved. This is not rocket science, just swap the memory around to isolate the bad DIMM (or DIMMs).
Click to expand...

There is a memory problem (or appears to be), but it does not in any way rule out that there is a driver issue (corrupted, misconfigured, etc).

A separate issue is whether hardware could interfere with memtest via DMA, interrupts, etc, such that certain addresses need to be taken with a grain of salt. Memtest86 does have options to avoid certain areas of memory, and I have asked to it not to. I'll have to test with default settings as well.
 
DeaconFrost said:
You'd be speculating incorrectly then. It's a common sense troubleshooting step to disable automatic reboots in the event of a BSoD. I couldn't tell you when the last time I saw one a BSoD on a system I set up and/or built. I'm not sure how you could twist my words into me somehow stating my systems were unstable from that. XP's been out for 8 years now or so, and from the very beginning, this is one of the very first configuration changes most would change. How else would you ever troubleshoot a BSoD if you encountered one?
Click to expand...

You could look in event viewer from the control panel on the next reboot. assuming your system isn't in a perpetual reboot cycle.
 
The final update is that one stick consistently shows errors, albeit different errors in different (adjacent or nearlby) locations. Crucial has been contacted for RMA.

Replacing the sticks with the old ram (ocz 2x1gb) seems to have eliminated the problem so far. I'm still wondering why the blue screen showed consistently the same error (same file, same address) since there's little guarantee that the file would be loaded to the exact same address on each boot. The driver could have been referring to that location, but I have no good way to figure out what that address was allocated for or mapped to.
 
venm11 said:
Review your first post and explain how that statement could help this issue. The tips you are giving are simply not of any use, we're way beyond that. Please... enough.
Click to expand...
They didn't need to be useful, because you already had your answer. Your reply to me goes against your very own advice. So the "please....enough" comment can very easily be directed at you as well. We have enough hypocrites already, who like to decry others while doing the very same thing themselves. If you were so high and almighty to put down others and lecture on pontificating, you would have known about this simple setting. You didn't, it got you riled up, and you lashed out. Seen it a million times. As one great poster used to say.......next!
 
Memory errors or a bad OC can cause all sorts of weird BSODs. But IRQ_LESS_THAN_EQUAL shows up a ton during overclocks. I agree with ryan - check eventvwr << enter that into Run.

Just look for the red X's under System.
 
Strange... I edited post 21 as the BSOD is actually still occurring; changing the bad memory out didn't help.

As posted earlier, there's nothing in the event log. I don't think it is able to write the event log when this error occurs. I could try reinstalling drivers, but I'm not able to find that file in the nvidia driver pack or on the disk, which is odd.
 
venm11 said:
Strange... I edited post 21 as the BSOD is actually still occurring; changing the bad memory out didn't help.

As posted earlier, there's nothing in the event log. I don't think it is able to write the event log when this error occurs. I could try reinstalling drivers, but I'm not able to find that file in the nvidia driver pack or on the disk, which is odd.
Click to expand...

What chipset are you running on?
 
DeaconFrost said:
I am truly amazed, here in 2009, that this isn't one of the first things EVERYONE does to their systems (at least if they still have XP).
Click to expand...
Yea, well most of the time when someone sees an error, they blow right by it. I don't know how many times I've asked people to tell me what the error said - or if they wrote it down.......... and the answer's been "I don't know", or "no I didn't".
 
venm11 said:
nforce4 ultra
Click to expand...

Are you using the ActiveArmor feature? If so, turn it off. It is known to cause problems due to hardware level bugs.

Otherwise, make sure you have the latest nForce dirvers installed. NV's IDE drivers have quite a history of poor quality and searching your error on Google turn up three posts from various years with the exact same BCCode parameters you've listed. If you still can't fixed the problem with a reinstall of the drivers, it's probably best if you don't use the NV IDE drivers at all.
 
Ryan, if you have those links handy, I'd be interested in reading them.

The latest nforce drivers had actually been installed for awhile when this started happening. I realized this when downloading the "latest" ones. I'll have to try and reinstall them after xmas. What bugs me is that I can't find that specific file anywhere. Also, I'd rather stick with nvidia's ata drivers because sometimes there are performance improvements over the default windows drivers.

And no, I don't have activearmor installed. It seems redundant.
 
Last edited:
venm11 said:
Ryan, if you have those links handy, I'd be interested in reading them.

The latest nforce drivers had actually been installed for awhile when this started happening. I realized this when downloading the "latest" ones. I'll have to try and reinstall them after xmas. What bugs me is that I can't find that specific file anywhere. Also, I'd rather stick with nvidia's ata drivers because sometimes there are performance improvements over the default windows drivers.

And no, I don't have activearmor installed. It seems redundant.
Click to expand...

just google "nvata.sys bsod" and you'll see plenty of links.
 
Well, I uninstalled the nvidia drivers, and that turned out to not be a good thing. A backup would have been smart. It would appear that the windows default drivers aren't working so well. Explorer seems to be non-functional.
 
Boot to a liveCD so that you can access your main system drive. Browse to your Windows folder and in there there'll be a Minidump folder. Copy that whole folder to a flash drive or whatever. On a different computer, download and install the Windows Debugging Tool. http://www.microsoft.com/whdc/DevTools/Debugging/default.mspx
Open that and open the newest Minidump from that file. Near the bottom it should tell you what most likely caused the BSOD.

That tool has helped many times in tracking down the culprit of a BSOD.
 
/usr/home said:
Boot to a liveCD so that you can access your main system drive. Browse to your Windows folder and in there there'll be a Minidump folder. Copy that whole folder to a flash drive or whatever. On a different computer, download and install the Windows Debugging Tool. http://www.microsoft.com/whdc/DevTools/Debugging/default.mspx
Open that and open the newest Minidump from that file. Near the bottom it should tell you what most likely caused the BSOD.

That tool has helped many times in tracking down the culprit of a BSOD.
Click to expand...


He most likely won't be able to get a minidump since it's his storage driver that's causing the BSOD.
 
/usr/home --- thanks for the tip on the debugging tool, that's good to have.

So... booting in safe mode allowed me to reinstall the drivers and everything seems to be working fine. I was able to simultaneously defrag and burn a dvd, either of which would crash it previously. Apparently I did not have the current nvidia drivers (6.xx specifically for nforce4ultra) and what I installed appears to be differently organized (15.xx for general nforce).

Anyway, thanks for your help, guys.
 
You must log in or register to reply here.
Back
Top