Questions

Computer reboots unexpectedly and applications err frequently

+
0 Votes
Locked

Computer reboots unexpectedly and applications err frequently

jon_melvin
I am technically savvy, though baffled. I designed and built a computer system selecting the individual components.

Recently, my computer system has been randomly rebooting (Windows XP Pro SP2). Additionally, applications ranging from GAIM, to Firefox to Neverwinter Nights are crashing with application errors. The modules causing these errors are sometimes the application executable itself, other times a supporting DLL.

More often than not, the memory offset for these errors is a low value.

Recently, I had to replace the northbridge stock fan/heatsink on the DFI NF4 SLI-DR with a Jingting Offset Cooler. Chipset temperatures are much cooler, averaging around 42C at load vs. 54C.

Any ideas what might be causing these problems? This forum is my last recourse before a wipe and reinstall. Some of my application errors
are posted below. I also have Minidumps available, though I cannot make sense of them.

Any help would be greatly appreciated. Thank you!

-Jonathan

Faulting application firefox.exe, version 1.8.20061.20418, faulting
module js3250.dll, version 4.0.0.0, fault address 0x0001eaaa.

Faulting application gaim.exe, version 2.0.0.0, faulting module
libpango-1.0-0.dll, version 1.14.5.0, fault address 0x000126ce.

Faulting application firefox.exe, version 1.8.20061.20418, faulting
module js3250.dll, version 4.0.0.0, fault address 0x0001f728.

Faulting application explorer.exe, version 6.0.2900.2180, faulting
module comctl32.dll, version 6.0.2900.2982, fault address 0x000730b0.

Faulting application nwn2main.exe, version 1.0.0.0, faulting module
nwn2main.exe, version 1.0.0.0, fault address 0x003ca090.

Faulting application firefox.exe, version 1.8.20061.20418, faulting
module ntdll.dll, version 5.1.2600.2180, fault address 0x00012131.

Faulting application firefox.exe, version 1.8.20061.20418, faulting
module ntdll.dll, version 5.1.2600.2180, fault address 0x00011dac.

Faulting application firefox.exe, version 1.8.20061.20418, faulting
module npswf32.dll, version 9.0.28.0, fault address 0x000a3919.
  • +
    0 Votes
    TheChas

    Usually, sudden reboots are caused by 1 of 3 sources:

    CPU

    Motherboard

    Power Supply

    Did the chip-set fan fail? Or, did you replace it just for performance?

    Before wiping the hard drive and performing a reload, I would at least swap in a different power supply.

    My friends and I have gone through a number of power supplies lately. The under $20 500 and 600 watt supplies just are not holding up as well as the older $30 300 watt supplies used to.

    Chas

    +
    0 Votes
    HAL 9000 Moderator

    I would recommend a 340 W Antec True Power Supply to be fitted and then run a bunch of Diagnostic Utilities off the Ultimate Boot Cd available for download here

    http://tinyurl.com/3jnpy

    Run the RAM, M'Board and CPU Tests most of which are never ending loops so you can run till it drops or just continues to work.

    The reason that I recommend the Antec PS's is because they are good and rated to a continuous Power Rating not a Peek Value so they just work and as Power Supplies are one of the most overlooked items in any computer they often cause a lot of problems that are attributed to other things. Besides swapping out a PS is a lot faster and less expensive if your time is taken into account than a reload.

    You might like to try one of the many available Live Linux's to see if the platform behaves itself with a different OS running. The Live Linux runs off a CD or DVD and Loads in RAM so it doesn't affect any already installed OS but it gives a good indication if you have a software problem rather than a Hardware problem. A fast and easy test with very little work involved. You can either download a Live Linux or buy one from here

    http://tinyurl.com/23nmq

    According to the Web Site they sell these from 99 Cents US with 49 cents US postage so this may be much cheaper to buy than to download and The Ultimate Boot CD is also available here now as well.

    If the system works perfectly with a Live Linux running it you have a software problem but if it falls over like it is with Windows you know that the problem is Hardware related.

    Col

    +
    0 Votes
    IT cowgirl

    This is some of the best advice I have seen. So simple, yet so informing. It also helps to pinpoint the issue faster.

    Although I knew these were available, I never bothered to try them for these vague issues.

    Thanks!

    +
    0 Votes
    IT cowgirl

    I agree with both TheChas and HAL9000.

    A faulty memory module can cause lots of application errors and some reboots. But your issues sound more like cpu or power supply.

    Good Luck! Let us know the outcome.

    +
    0 Votes
    willcomp

    I agree with Colin (Hal9000) that booting from a live Linux CD is an excellent way to differentiate between hardware and software issues.

    What is make/model/wattage of power supply?
    CPU?
    Video card(s)?
    Hard disks?
    Optical drives?

    Have you recently updated or changed any major drivers (e.g. video or chipset)?

    As previously mentioned, the major hardware suspects are power supply, memory, CPU, and motherboard.

    Could also be a driver issue or inadequate power supply. Based on your mobo, I would expect you need a quality 600 watt or better SLI rated PSU.

    +
    0 Votes
    jon_melvin

    Thank you so much for all the great information and suggestions thus far. I really hope its not a hardware issue. I will give the Linux Live CD deal a try, keeping my fingers crossed.

    When I replaced the stock chipset cooler/fan on my DFI LP NF4 SLI-DR mobo, it was going haywire--bearing was busted or something. It was groaning, whining and making 747 takeoff sounds. I will let everyone know how my Linux tests go. Thanks again, and if you can think of anything else, please post. Have a great day/weekend!

    My system specs are below:

    AMD Athlon 64 4000+ San Diego (@ 2.4Ghz)

    ZALMAN CNPS7000B-CU PURE COPPER BASE CPU COOLING FAN

    Corsair TWINX1024-4400C25 1GB DDR KIT

    SEAGATE 300GB ST3300831AS SATA

    CREATIVE LABS SOUNDBLASTER AUDIGY2 ZS PLATINUM

    ATI T550PRO TV TUNER

    eVGA GEFORCE 6800GT 256MB PCI EXPRESS

    Antec Neo 600W PS


    -Jonathan

    +
    0 Votes
    willcomp

    That PSU should be sufficient. So don't believe you have an underpowered system.

    Let us know how Linux testing goes.

    Any driver updates/changes? You may want to try another video driver. If you have latest installed, drop back a rev or 2. If older version installed, try latest.

    +
    0 Votes
    jon_melvin

    I have the most current non-beta version of NVIDIA's video drivers installed '93.71' I'll try regressing driver versions.

    This morning, 4 passes of MEMTEST were run with no errors or reboots.

    I'll run the system burn-in test "Lucifer" overnight on the system to see how that goes.

    +
    0 Votes
    PC Services

    Have you checked "Services"? Control Panel>AdminTools>Services>"Remote Procedure Call (RPC)Locator"(click on to Services) click the Recovery tab, check to see if the response services tab is set to "Take No Action". There is a virus that will move in and change the computers response settings. This can be really irritating. However, normally only the OS will reboot with the Virus I have seen.

    Did you clone your OS with Ghost?
    If so it might be a Power Management problem, designation and source Power Management are different.
    Or a Disk Signature issue if the disk did not reinitialize when restoring the partition.

    +
    0 Votes
    jon_melvin

    The notion I would have to replace hardware may very well hold true.

    Lucifer ran overnight (8 hours) and encountered 26 errors, though continued running.

    The CPU tests would complete iterations, though within a hours times, the screen would go blank--seemingly with the system hung.

    Memtest has run without any errors and 20 full cycles.


    With all this said and done, is there any way to know definitely what hardware component may be failing me?

    I find it more than a little odd that most all of the application errors mentioned at the top of this thread seem to have closely grouped memory offsets. Any input would be greatly appreciated. Thanks much!

    +
    0 Votes
    cleaverx

    All the advice so far has been right on, but just in case, you should run chkdsk /r from the recovery console. After that, make sure your page file size is set to system managed. Also I would run the system with a different PS (Buy one, try it, return it if no changes), and ultimately, remove and re-seat all adapters, memory, cpu and chipset heatsinks (these last two are of course critical). Keep us posted.
    CleaverX

    +
    0 Votes
    Conundrum

    I used to have similiar prblems. The cause of my random reboots turned out to be a perfomce setting in my Nvidia 6600. It was set to quality, I set it to performance, no reboots for a while. This may or may not help you.

  • +
    0 Votes
    TheChas

    Usually, sudden reboots are caused by 1 of 3 sources:

    CPU

    Motherboard

    Power Supply

    Did the chip-set fan fail? Or, did you replace it just for performance?

    Before wiping the hard drive and performing a reload, I would at least swap in a different power supply.

    My friends and I have gone through a number of power supplies lately. The under $20 500 and 600 watt supplies just are not holding up as well as the older $30 300 watt supplies used to.

    Chas

    +
    0 Votes
    HAL 9000 Moderator

    I would recommend a 340 W Antec True Power Supply to be fitted and then run a bunch of Diagnostic Utilities off the Ultimate Boot Cd available for download here

    http://tinyurl.com/3jnpy

    Run the RAM, M'Board and CPU Tests most of which are never ending loops so you can run till it drops or just continues to work.

    The reason that I recommend the Antec PS's is because they are good and rated to a continuous Power Rating not a Peek Value so they just work and as Power Supplies are one of the most overlooked items in any computer they often cause a lot of problems that are attributed to other things. Besides swapping out a PS is a lot faster and less expensive if your time is taken into account than a reload.

    You might like to try one of the many available Live Linux's to see if the platform behaves itself with a different OS running. The Live Linux runs off a CD or DVD and Loads in RAM so it doesn't affect any already installed OS but it gives a good indication if you have a software problem rather than a Hardware problem. A fast and easy test with very little work involved. You can either download a Live Linux or buy one from here

    http://tinyurl.com/23nmq

    According to the Web Site they sell these from 99 Cents US with 49 cents US postage so this may be much cheaper to buy than to download and The Ultimate Boot CD is also available here now as well.

    If the system works perfectly with a Live Linux running it you have a software problem but if it falls over like it is with Windows you know that the problem is Hardware related.

    Col

    +
    0 Votes
    IT cowgirl

    This is some of the best advice I have seen. So simple, yet so informing. It also helps to pinpoint the issue faster.

    Although I knew these were available, I never bothered to try them for these vague issues.

    Thanks!

    +
    0 Votes
    IT cowgirl

    I agree with both TheChas and HAL9000.

    A faulty memory module can cause lots of application errors and some reboots. But your issues sound more like cpu or power supply.

    Good Luck! Let us know the outcome.

    +
    0 Votes
    willcomp

    I agree with Colin (Hal9000) that booting from a live Linux CD is an excellent way to differentiate between hardware and software issues.

    What is make/model/wattage of power supply?
    CPU?
    Video card(s)?
    Hard disks?
    Optical drives?

    Have you recently updated or changed any major drivers (e.g. video or chipset)?

    As previously mentioned, the major hardware suspects are power supply, memory, CPU, and motherboard.

    Could also be a driver issue or inadequate power supply. Based on your mobo, I would expect you need a quality 600 watt or better SLI rated PSU.

    +
    0 Votes
    jon_melvin

    Thank you so much for all the great information and suggestions thus far. I really hope its not a hardware issue. I will give the Linux Live CD deal a try, keeping my fingers crossed.

    When I replaced the stock chipset cooler/fan on my DFI LP NF4 SLI-DR mobo, it was going haywire--bearing was busted or something. It was groaning, whining and making 747 takeoff sounds. I will let everyone know how my Linux tests go. Thanks again, and if you can think of anything else, please post. Have a great day/weekend!

    My system specs are below:

    AMD Athlon 64 4000+ San Diego (@ 2.4Ghz)

    ZALMAN CNPS7000B-CU PURE COPPER BASE CPU COOLING FAN

    Corsair TWINX1024-4400C25 1GB DDR KIT

    SEAGATE 300GB ST3300831AS SATA

    CREATIVE LABS SOUNDBLASTER AUDIGY2 ZS PLATINUM

    ATI T550PRO TV TUNER

    eVGA GEFORCE 6800GT 256MB PCI EXPRESS

    Antec Neo 600W PS


    -Jonathan

    +
    0 Votes
    willcomp

    That PSU should be sufficient. So don't believe you have an underpowered system.

    Let us know how Linux testing goes.

    Any driver updates/changes? You may want to try another video driver. If you have latest installed, drop back a rev or 2. If older version installed, try latest.

    +
    0 Votes
    jon_melvin

    I have the most current non-beta version of NVIDIA's video drivers installed '93.71' I'll try regressing driver versions.

    This morning, 4 passes of MEMTEST were run with no errors or reboots.

    I'll run the system burn-in test "Lucifer" overnight on the system to see how that goes.

    +
    0 Votes
    PC Services

    Have you checked "Services"? Control Panel>AdminTools>Services>"Remote Procedure Call (RPC)Locator"(click on to Services) click the Recovery tab, check to see if the response services tab is set to "Take No Action". There is a virus that will move in and change the computers response settings. This can be really irritating. However, normally only the OS will reboot with the Virus I have seen.

    Did you clone your OS with Ghost?
    If so it might be a Power Management problem, designation and source Power Management are different.
    Or a Disk Signature issue if the disk did not reinitialize when restoring the partition.

    +
    0 Votes
    jon_melvin

    The notion I would have to replace hardware may very well hold true.

    Lucifer ran overnight (8 hours) and encountered 26 errors, though continued running.

    The CPU tests would complete iterations, though within a hours times, the screen would go blank--seemingly with the system hung.

    Memtest has run without any errors and 20 full cycles.


    With all this said and done, is there any way to know definitely what hardware component may be failing me?

    I find it more than a little odd that most all of the application errors mentioned at the top of this thread seem to have closely grouped memory offsets. Any input would be greatly appreciated. Thanks much!

    +
    0 Votes
    cleaverx

    All the advice so far has been right on, but just in case, you should run chkdsk /r from the recovery console. After that, make sure your page file size is set to system managed. Also I would run the system with a different PS (Buy one, try it, return it if no changes), and ultimately, remove and re-seat all adapters, memory, cpu and chipset heatsinks (these last two are of course critical). Keep us posted.
    CleaverX

    +
    0 Votes
    Conundrum

    I used to have similiar prblems. The cause of my random reboots turned out to be a perfomce setting in my Nvidia 6600. It was set to quality, I set it to performance, no reboots for a while. This may or may not help you.