Wndows XP Random Reboots

[Coldfinger67]

I recently had my PC rebuilt (August '06) from a Pentium 4 1.8 ghtz CPU to this machine

Intel Pentium D 930 CPU @ 3 ghz
Microstar MB Model PM8M3-V (VH)
1 gb pc3200 ddr400 RAM
600 watt OCZ Tech ATX PSU
XFX GeForce 6800 (AGP) with driver 91.47OS
Windows XP – SP2 with all updates to date

These were carried over from original machine:

WD HDDs - 120 gig and 30 gig (for backups, etc)
Sound Blaster Live 5.1
US Robotics 3095 Modem
3.5" floppy drive
CD-Rom Read Only drive
DVD-Rom Read/Write drive

Original build in August 2006 also had Nvidia GeForce 5500 with driver 84.21
And 400 watt ATX generic which were swapped out the begining of this month.

System random reboots started within a week or so of the rebuild – PSU and video card upgraded but no change with reboots continuing

All peripherial divers updated to latest versions
Award MB BIOS flashed on build – ver 4.3

Majority of reboots have occurred while in game play (Max Payne 2 & Hitman Blood Money). However have also occurred while using media player 10 and in IE doing websearches about random reboots, etc.

All system error reboots have exactly the same code:

Bugcheck 0x0000009c Parameters 0x00000000 – 0xf7937050 – 0xb2000000 – 0x1040080f

Sound card and modem were removed with no change in reboot pattern.

Gig stick removed, contacts cleaned, sockets blown clean with compressed air and reseated in second slot and then checked with latest version of memtest 86 with three separate passes - no errors detected. WD Diagnostics show nothing with main HDD.

I have gone into BIOS on several of the reboots and the CPU temp is showing right around 50 degrees c - well within operating parameters so I do not think that overheating is an issue. ALthough I unchecked the automatic reboot option in Administrative Tools, the PC continues to reboot with no BSOD coming up. Since the reboots started, only one BSOD has come up and that one indicated something recently installed was having a problem but this message came up within a week or two of the rebuild so everything had just been installed so I did not see anything at that time as helpful.

NOTE: this is probably a second unconnected problem that recently cropped up.

In Security Event Log, Windows Defender Id’s a “sys” file in Docu~\User~\Local ~\Temp named “o1394bul.sys” as a potential spyware threat. Event Log states WinDef takes necessary steps to protect system against threat. However, this file name comes back from a worldwide websearch with only foreign (non-english) references. Windows XP and MS data bases return no hits either. Machine has begun to attempt to establish a dial-up connection all by itself over the last few days.

I have the minidump files available to email or post if that would be helpful. From what I have been able to tell from the research I have done is that this particular code "9c" most likely references a hardware problem. Since the RAM stick passed memtest 86, the reboots happen with the sound and modem cards removed and the video card was upgraded with a different driver and with reboots continuing, it would seem that attention should be focused to the MB and or CPU???

Any thoughts or suggestions would be greatly appreciated as I have to build a case to present to the local distributor's tech staff if the MB or CPU needs to be replaced.

Thanks for any help

 

[Rik]

Try getting rid of windows defender, its not a very good bit of software.
Its possible just doing that may fix your problem.

 

[Tmagic650]

Try getting rid of windows defender, its not a very good bit of software.
Its possible just doing that may fix your problem.

I agree

Coldfinger67,
go into Control Panel, System, Advanced, Start Up and Recovery, Settings, and un-check automatically restart. This may force Windows to report an error on the screen, that will give us a better clue as to what is happening

 

[Coldfinger67]

First, thanks for the quick replies. PC came back to me with WinDef (Beta) already installed. Figured it was included automatically in the latest version of SP-2??? Agree it is seemingly pretty bad spyware at the moment. I am certainly "no pro from dover" but given the error codes I can't see the link. Will look to see if it is removable under "Add/Remove".

Second, strangely enough I did uncheck the box early in the problem and have checked several times since and it is still unchecked but the system continues to auto reboot itself, thus cutting out the chance for any BSOD????

 

[Tmagic650]

Ok Coldfinger67,
it could be that your memory timings are a bit to tight. Try backing off the timings or set them to the bios defaults and see if the system stabilizes

 

[Coldfinger67]

Please note that I am not a "pro from dover" when it comes to changing any settings in BIOS. Since I got the machine back, I have not tweaked any settings and I doubt the guy who did the rebuild did either. I have simply used it to monitor CPU temps as MSI tells me my MB will not support independent monitoring software such as PCAlert.

If you want me to check what the setting is, some direction would be needed. Someone had suggested kicking up the voltage to the RAM to 2.7 but had no idea and could not find in the BIOS and settings that would allow that.

 

[Tmagic650]

This could be caused by too low memory voltages... You came here for help, so that's what we are trying to provide. If you aren't comfortable with trying to adjust these timings or voltages, try to get the guy that built your system to tweak them for you

 

[Coldfinger67]

I am certainly more than appreciative of any help with this problem. I was simply acknowledging that most responders on this forum are likely to be much better and smarter at this than I am and may offer suggestions I do not quickly understand and simply need somethings "dumbed down" to my level. I am not necessarily uncomfortable with tweaking BIOS just want to make sure I am doing the right thing and do not make a bad matter worse.

When the advice was made to increase the voltage to the RAM, I went into BIOS, through every single entry and could not find any one that appeared would allow voltage to be adjusted manually...at least not that I could tell....might it say something besides RAM voltage adjustment???? I am assuming that most bios settings are standard??? There was only one that looked remotely like it could be a voltage adjustment but it was a toggle set at 2.5 and the next toggle was 3 (vs. the advice was to bump it up to only 2.7) so I was unsure exactly what that setting did and what might happen if I changed the value from 2.5 to 3.

Again I am most greatful for any help with this....thanks for the reply...

 

[countrywill]

I recently had my PC rebuilt (August '06) from a Pentium 4 1.8 ghtz CPU to this machine

Intel Pentium D 930 CPU @ 3 ghz
Microstar MB Model PM8M3-V (VH)
1 gb pc3200 ddr400 RAM
600 watt OCZ Tech ATX PSU
XFX GeForce 6800 (AGP) with driver 91.47OS
Windows XP – SP2 with all updates to date

These were carried over from original machine:

WD HDDs - 120 gig and 30 gig (for backups, etc)
Sound Blaster Live 5.1
US Robotics 3095 Modem
3.5" floppy drive
CD-Rom Read Only drive
DVD-Rom Read/Write drive

Original build in August 2006 also had Nvidia GeForce 5500 with driver 84.21
And 400 watt ATX generic which were swapped out the begining of this month.

System random reboots started within a week or so of the rebuild – PSU and video card upgraded but no change with reboots continuing

All peripherial divers updated to latest versions
Award MB BIOS flashed on build – ver 4.3

Majority of reboots have occurred while in game play (Max Payne 2 & Hitman Blood Money). However have also occurred while using media player 10 and in IE doing websearches about random reboots, etc.

All system error reboots have exactly the same code:

Bugcheck 0x0000009c Parameters 0x00000000 – 0xf7937050 – 0xb2000000 – 0x1040080f

Sound card and modem were removed with no change in reboot pattern.

Gig stick removed, contacts cleaned, sockets blown clean with compressed air and reseated in second slot and then checked with latest version of memtest 86 with three separate passes - no errors detected. WD Diagnostics show nothing with main HDD.

I have gone into BIOS on several of the reboots and the CPU temp is showing right around 50 degrees c - well within operating parameters so I do not think that overheating is an issue. ALthough I unchecked the automatic reboot option in Administrative Tools, the PC continues to reboot with no BSOD coming up. Since the reboots started, only one BSOD has come up and that one indicated something recently installed was having a problem but this message came up within a week or two of the rebuild so everything had just been installed so I did not see anything at that time as helpful.

NOTE: this is probably a second unconnected problem that recently cropped up.

In Security Event Log, Windows Defender Id’s a “sys” file in Docu~\User~\Local ~\Temp named “o1394bul.sys” as a potential spyware threat. Event Log states WinDef takes necessary steps to protect system against threat. However, this file name comes back from a worldwide websearch with only foreign (non-english) references. Windows XP and MS data bases return no hits either. Machine has begun to attempt to establish a dial-up connection all by itself over the last few days.

I have the minidump files available to email or post if that would be helpful. From what I have been able to tell from the research I have done is that this particular code "9c" most likely references a hardware problem. Since the RAM stick passed memtest 86, the reboots happen with the sound and modem cards removed and the video card was upgraded with a different driver and with reboots continuing, it would seem that attention should be focused to the MB and or CPU???

Any thoughts or suggestions would be greatly appreciated as I have to build a case to present to the local distributor's tech staff if the MB or CPU needs to be replaced.

Thanks for any help[/QUOTE

My PC was doing the same thing and sometimes would not boot at all. The problem was the CPU was over heating and cutting out This was also causing the capacitors at the side of the CPU to swell on the top.
The problem was solved by cleaning the heat sink on the CPU and re-newing the heat sink compound under the heat sink.

Hope this helps
Lianne

 

[Sunsetdrunk]

Have you narrowed it to software or hardware?

I'm missing something. Did a rebuild include a fresh drive and Windows XP install?

 

[Tmagic650]

I am certainly more than appreciative of any help with this problem. I was simply acknowledging that most responders on this forum are likely to be much better and smarter at this than I am and may offer suggestions I do not quickly understand and simply need somethings "dumbed down" to my level. I am not necessarily uncomfortable with tweaking BIOS just want to make sure I am doing the right thing and do not make a bad matter worse.

When the advice was made to increase the voltage to the RAM, I went into BIOS, through every single entry and could not find any one that appeared would allow voltage to be adjusted manually...at least not that I could tell....might it say something besides RAM voltage adjustment???? I am assuming that most bios settings are standard??? There was only one that looked remotely like it could be a voltage adjustment but it was a toggle set at 2.5 and the next toggle was 3 (vs. the advice was to bump it up to only 2.7) so I was unsure exactly what that setting did and what might happen if I changed the value from 2.5 to 3.

Again I am most greatful for any help with this....thanks for the reply...

I've been doing some more thinking on this... Why have you come here, instead of taking your system back to the builders and let them get the "bugs" out of it? August '06, it's less than a month since you took delivery right? Have you done something that might have voided your warranty? I build systems and I give a 90-day on my work, 6 months for parts...

 

[Coldfinger67]

The machine was built by a close friend of mine who does this as a second job. Unfortuanetly, He has been having a number of personal things to deal with came up on him since the rebuild so I told him I would do some research for him rather than have my machine sit on his worktable until he can get other things taken care of. He is very knowledgeable but he would have to be doing the same things I am trying before the distributor he deals with will simply turn over replacement parts without knowing which component is the bad one. Again the machine is fully warranted by him and the parts are warranted by the manufacturer but we have to find out what part is bad. Hence my efforts to run the machine with different components out of the system to try and narrow down what is happening. From what I read on previous threads about this issue on this site, it seemed to me that someone here might be able to shed some light on a system error that continues to be exactly the same each time. I am not able to read a minidump nor do I have the tools to try. Was hoping again someone could at least point me in the right direction so that when my friend is able to devote some time to it, I will have at least done some of the leg work for him.

I actually installed the new PSU and the video card and have installed all the new updated drivers. So I have a fair idea of what I am doing but the reboots were going on before and now after these attempts to rectify the possible source of the problem. So I am near the end of my knowledge base and hoped to reach out to others.

If you really want to help, stop being so suspicious and try and help me figure out what the stop code is pointing to. An earlier thread that brought me to this site in the first place by one of the senior techs from this site (Howard) indicated that a "9c" was most likely hardware related (CPU, RAM or MB) and more might be determined by examining the minidumps. I will be happy to post them or email them if you or anyone else can help read them and decipher whether they hold any clues to what is causing the reboots.

I did not appreciate your accusatory tone. I am a senior level official with the federal government and IT is something I have had involvement with but more on the software side vs. hardware. I was hoping this site was populated by people willing to lend their expertise to solve problems and offer useful advice to someone like me trying to learn something new, expand my knowledge base and hopefully at the same time maybe helping my friend who has his own other problems right now.

So if you have something constructive and informative to offer me, please do if you so choose. After all this is America and you have a choice.

 

[Tmagic650]

I appologise for my tone. You are dealing with a fairly difficult problem, only because there can be many causes for it. Memory timings, and voltages are very tricky and should only be changed by the qualified... CPU overheating is tricky too. If the thermal paste was improperly applied or the HSF CPU Heatsink and Fan improperly fitted, unless you are familiar with doing this, I wouldn't attempt messing with it either.

Where does that leave us? Bad memory, you can replace. Resetting the bios to default settings, you can do this, right?

I just built an Intel Celeron D 2.53GHz system on an MSI PM8M3-V, 1GB RAM, Seagate 250GB SATA, Sapphire Radeon 9600 256MB AGP 8X card...

To adjust memory timings, and voltage go to Advanced Chipset Features, DRAM/Clock/Drive Control. You can change DRAM Clock, and DRAM Timing. If you change DRAM Timing to MANUAL, you can change all those things that you didn't see before. Please write down the settings before you change anything! Play with settings. If you make a mistake, all you have to do to get the computer to boot again, is to reset the CMOS. It is easy enough

 

[Rick]

Your dumps all blame NTOSKRNL.EXE, which is the system kernel. This program is the heart of Windows. It shares your processor, manages memory, handles communications and much, much more. So this doesn't really help us, because it could be just about anything. This tells us that elimination is probably going to be the most practical way to troubleshoot.

From my experiences, simliar errors have always been a result of hardware failure. The random nature of the error suggests strongly that your memory, CPU, power supply and/or system board are the problem. (In order of likelyhood). Tmagic650 is on the money with the suggestions provided.

I feel very strongly that memory is the most likely culprit. It could also very well be a defective, overheating or excessively overclocked processor/system board or even a failing power supply.

You should have the memory swapped out as part of the elimination process.

To verify my memory hunch, you can run memtest+ ( www.memtest.org ). Download the floppy image and create a bootable floppy disk or download the pre-compiled ISO image and burn it with a program like Nero to make a bootable CD. Let it run at least overnight, although it is very likely you'll get errors in the first 5 minutes. If the problem you're experiencing happens frequently, but you can run memtest for hours on end without issues, then we can start thinking about software problems.... Baby steps though.

Testing the memory will reveal memory errors, CPU errors and possibly board problems as well as overheating. But more often than not, memory errors are indeed caused by bad memory.

Best of luck.

 

[Tmagic650]

Right Rick,
memory replacememt would be my first chioce. I have used memtest, but I have found that it either shows errors quickly or not at all... and when not at all, it doesn't mean that the memory is good. If you are running 2 sticks, only 1 will be bad. I have never seen 2 bad sticks on the same board ever!

It's much faster just to replace suspect memory. Afterall, time is money

 

[Coldfinger67]

Thanks for the positive response. In life, especially when conversing like this, no one really knows who anyone is but try not to assume the worse in people. There are some good people out, yourself included, whose motiviations are not always apparent. In my case I am simply trying to learn something and at the same time trying to reduce what is going to obviously be some bench time.

I finally had a chance to sit with my friend this afternoon and he is equally perplexed. He has been doing builds for me for more than 15 years and I particularly seem to end up with the most unusual problems. He has not had problems like this random reboot issue in quite some time. He generally leaves most of the BIOS settings at the defaults and he says he has never had to overclock or change voltage or timing on DRAM and would be concerned that at this stage it could void the warranty on the parts.

Here is where I am at. I let memtest86 run all night long. It made over 18 passes and absolutely no errors. The CPU and heat sink are clearly well pasted together. I attempted to remove it to check on the status of the transfer medium but no way without removing the motherboard. He said he used the heat transfer material than came with the intel chip and heatsink and fan.

It did another reboot last night during the short time I had to play. Same Stop error and parameters.

I guess one thing I did not understand was the comment by Rick that my minidumps pointed to a particular exe file? I have offered but have not posted any of the minidumps. Did he mean to say the stop error and parameters indicated that??? Again I will be happy to post either a zip with all of the ones still retained or just the last few???

My friend will be in better shape next week to give the machine back to him. His plan of attack at the moment is to probably remove the MB with the CPU and RAM and take it to his distributor who has the bench ability to test the components thoroughly and in fact require it in order to get them to replace the items.

I will check with him tomorrow and see if he has another 1 gig memory stick that I could swap out over the weekend and see if that makes any difference. By he may not. The only stick I have around is a 512 that came out of the last version of my machine. Not sure how XP might react to having a different stick in with perhaps different characteristics.

But please do not give up on offering suggestions as I am pointing him here to see what others in the same field have to say.

Again thanks....

 

[Rik]

First off, get speedfan from the downloads section, its free and will monitor fan-speeds, voltages, and temperatures. It should work fine with your mobo as i used to use an MSI. I cant absolutely guarantee that it will work as i used to have an MSI 745 Ultra but its worth a try.

Windows Defender isn't an integral part of windows update and is a separate program and will uninstall fine. As an example, i used to use it until it decided my Internet dialler program was a threat and removed it. I then got etrust pest patrol which found over 20 bits of spyware even tho defender said my system was clean.

 

[Coldfinger67]

There seems to be some confusion at MSI - one of their tech support staff told me to download their PCAlert and use it to monitor system. Called back on a related matter and second tech told me that this particular MB did not support software monitoring and that was why PCAlert was not listed as a download under this particular MB.....so I got to supposed experts arguing about this..leaving this hapless guy asking the question shaking his head.

I suppose it could not hurt to load their software, worst case scenario it simply does not work

Here is a minidump zip of the last four that go with the previously posted Stop Error.....can anyone confirm Rick's observations as to what they indicate????

thanks again in advance

 

[Tmagic650]

Coldfinger67,
there is some confusion here. I just built a system using the same MSI motherboard as you have, and I assure you that MSI's Live Update, PCalert, and Fuzzy Logic will work with your system. You shoud have received a disc labeled "Drivers and Utilities", or your friend might have it. This disc has what you need. You can download MSI's Live Update from MSI's website. Run the utility and it will give you the opportunity to download Fuzzy Logic and PCalert. Fuzzy Logic displays information about your CPU and Memory settings and clocks. It will also adjust these speeds and settings automatically for optimum performance... and don't forget to replace that memory!

 

[Rick]

I guess one thing I did not understand was the comment by Rick that my minidumps pointed to a particular exe file? I have offered but have not posted any of the minidumps.

I don't know how, but I downloaded minidumps... I thought they were from this thread. Maybe I was looking at the wrong one.

Memtest runs overnight - this is a pretty decent indication that your core hardware components are stable. CPU, memory & system board are in good enough shape to run memtest overnight and should be good enough shape to run Windows for more than several minutes.

Have you tried a fresh install of Windows XP? I haven't read anything in this post that says for sure Windows has been installed from scratch on your new system.

Maybe a parallel installation on a seperate partition of hard drive could help give us more confidence this is a hardware or software problem. Your error is very 'clearly' hardware, but your troubleshooting seems to indicate that software is more likely.

 

[mecanic]

I would download some free burn-in software and install it, boot into safemode and run the burn-in over night. If it passes the stress test then
your problem would more than likely be a driver or software/xp related. My 1st guess would be the video card or the AGP chipset drivers....maybe try uninstalling the current video card driver and reinstalling a older version.

 

[Tmagic650]

There seems to be some confusion at MSI - one of their tech support staff told me to download their PCAlert and use it to monitor system. Called back on a related matter and second tech told me that this particular MB did not support software monitoring and that was why PCAlert was not listed as a download under this particular MB.....so I got to supposed experts arguing about this..leaving this hapless guy asking the question shaking his head.

I suppose it could not hurt to load their software, worst case scenario it simply does not work

Here is a minidump zip of the last four that go with the previously posted Stop Error.....can anyone confirm Rick's observations as to what they indicate????

thanks again in advance

Aren't there minidumps associated with this post? If I build a system for a customer, I always keep a copy of the installed OS in my possession. If I gave it to the customer, IT person or not, the chance of them messing up the system with it, is greatly increased

 

[peterdiva]

All the dumps have a bugcheck of 9C: Machine_Check_Exception, this is a hardware problem.
One is caused by mnmdd.sys - Frame buffer simulator (?).
Two are caused by intelppm.sys - Is this the processor driver?
And one is caused by ntkrnlmp.exe with a process of MDM.exe - this seems to be an IE debugger. http://www.liutilities.com/products/wintaskspro/processlibrary/mdm/.

I had the same error with my AMD CPU, I uninstalled the AMD processor driver and just used the Windows driver, and I haven't had a BSOD since.

 

[Coldfinger67]

Yes it was a clean install due to all the hardward being changed.

My builder is going to take the MB, CPU and Stick out of the case and take it to the distributor next week who has the bench ability to test the components thoroughly. If they pass then the next step would seem to be a fresh install of XP.

One interesting new twist. I switched out the 1 gig stick for the 512 that had been in the old machine. Reboots came fast and more frequent with that stick in. So swapped it back out again.....not sure what that might indicate......

 

[Coldfinger67]

The latest update is that the PC was put through 12+ hours of burn in software with nary a reboot....however, when out on web..you guessed it...reboot...

So at the moment have done a clean reinstall of XP Pro and did an image disk before adding back in any software beyond basic items......will begin to slowly rebuild software base and see what happens...next step is MB/CPU back to distributor with a request for replacement since every other component has been either switched out or disconnected with no change in reboot status....