Puzzled by XP Home crash

[mevensen]

I am cannot resolve my current PC crash, I was hoping for some insight.

To start:
I have Gateway system which has been rock solid for over two years. I have made no hardware changes in over six months. The only thing I've "installed" recently was a CruzerProfile USB Flash Drive (this has some software on the jump drive for encryption of file and an integrated biometric reader). To the best of my knowledge, nothing was actually installed on the system, but who knows?

The problem:
I started having random restarts yesterday. These were intially recovered with a reboot. The problem increased in frequency. It has now progressed to the point that I cannot reboot from the hard drive. I have attempted to reboot from the Windows CD and Norton SystemWorks Recovery Disk, both to no avail. When I do reboot, I get a variety of things happening, seemingly randomly:

1. The floppy drive grinds, the system beeps, and this keeps going until I power down. When this happens, I don't even get the prompts for access to BIOS or Boot Priority.

2. I actually get to the BIOS/Boot screen. If I do nothing (with a bootable CD in drive) sometimes it will try to access the CD, sometimes it will proceed to #1.

3. If, instead, I select the boot priority to the CD drive, I can get it to that point. Then, I will either get:
a. Blue screen stating "A problem has been detected and windows has been shut down to prevent damage to your computer." Followed by the line "PFN_LIST_CORRUPT", although once it said "IRQ_NOT_LESS_OR_EQUAL". This is followed by instructions to remove components and disable BIOS options.
b. A message stating "Si3114r5.sys is corrupted" there was another similar message, but I did not record that one.
c. A message stating something to the effect of WINNT\...\CONFIG.SYS is corrupt or absent.

4. I have tried following the instructions I saw on this forum with regards to the 3.c. message, but I can't even get to the restore console to try manipulating any of those files.

I can't think of any more information I can give to elucidate the problem further. Any help would be most appreciated. Obviously I want to preserve as much information, etc. as I can, but I'm not exactly optimistic at this point. Thanks in advance.

[glc]

Pull the hard drive and install it as an additional drive in another computer - and get your files backed up that way. You may wind up having to run a full system restore.

[DynamicTech]

I would start by testing the hardware with a self bootable diagnostic program.
I would guess that you probably have either a bad hard drive, bad memory, or possibly a bad cpu.
Sorry glc, I didn't realize you had posted.

[mevensen]

Originally Posted by glc Pull the hard drive and install it as an additional drive in another computer - and get your files backed up that way. You may wind up having to run a full system restore.

I was expecting something like that. Of course, now all I need is a second computer with a SATA controller

Actually, about 2 hours after posting, I decided to power up again, just to try, and it made it all the way to windows. Weird, huh?

Now, I'm wondering if this is just a lucky strike, and I'll need to do a fresh install, with perhaps a new hard drive. Or, if this was just a really weird bad patch. Any ideas?

[KilluminatiStyle]

Sounds to me like it's either a heat issue or a power supply issue. Both would cause random and down right weird stuff to happen.

[mevensen]

Originally Posted by KilluminatiStyle Sounds to me like it's either a heat issue or a power supply issue. Both would cause random and down right weird stuff to happen.

It may be heat, although I'd be suprised, I haven't changed any of the hardware in quite some time. What is the easiest way to tell if it's a heat issue?

As far as the power supply goes, I have a relatively new Antec supply (under 6 mos) that I upgraded to when I upped my graphics card. It hasn't given me problems before.

BTW, since it started working again, I ran what test programs I have (essentially CheckIt) on the memory and CPU without any issues arising.

[mevensen]

Some more info, not sure if it's pertinent.

I have gotten things running again, somewhat. I noticed that a number of my USB peripherals were not being recognized (two printers, PDA dock, wireless mouse). It appears that the hub I am using for these 4 devices isn't working (I reconnected the PDA dock and mouse directly to the system and got them recognized/functioning, but I don't have enough cabling or additional hub to do the same for the printers at the moment).

Could a problem with a USB hub create these errors? Or is this another clue as to what's going on with the system?

BTW, I did find info on Motherboard Monitor and the BIOS PC Health monitoring functions, but I am now at work, so I can't look into that until I get home tonight.

[glc]

If the USB hub is not AC-powered, it and its connected devices could be drawing more than the allotted 500ma from the USB port. This can damage the motherboard circuitry if it doesn't shut down to protect itself.

[mevensen]

Originally Posted by glc If the USB hub is not AC-powered, it and its connected devices could be drawing more than the allotted 500ma from the USB port. This can damage the motherboard circuitry if it doesn't shut down to protect itself.

It is a powered hub.

[mevensen]

Just an update, I'm still having problems, although I can now boot relatively reliably. Currently, I get a suddenly black screen, followed by an auto-reboot. After this, often times I will have to unplug and reconnect my USB mouse in order to get it working again.

I got Everest monitor and the Intel Active Monitor to try to help figure out what is happening. The CPU stays under 60 C, although it does run in the 50s pretty regularly under load. I noticed the GPU (6600 GT) ran hotter (50s to 60s).

I replaced my PSU (was an Antec 430W, now is Thermaltake 500W) to try to minimize any power issues. I had added a case fan to help with heat, and my graphics card stopped working, so I figured there might be a power problem going on.

I realize this is starting to move into a hardware instead of Windows topic (or is it?), but since I started here...

Does this sound like a CPU or motherboard issue? The minimal diagnostics I have (CheckIt, and the monitors in Everest, Sandra, and Intel software) read the test the CPU out as working, same with memory. I don't see similar tests for the motherboard itself.

Any input or assistance would be appreciated.

[GaryRouth]

Hi mevensen

glc's recommendation to backup data while you can is still a good idea!

Other troubleshooting you can try:
1) Start the system in Safe Mode. See if you have any problems with rebooting while running in Safe Mode for a somewhat extended period of time. Yes, your graphics won't look wonderful, but it can help with figuring whether this is a hardware or software problem.
2) Change the setting for Windows XP that causes the system to automatically reboot when encountering a fatal error. If a driver or other conflict is causing the problem, you aren't able to read the error message - unless you turn this behavior off. To do so, click on Start, then right-click on My Computer, then select Properties, then Advanced, then click the Settings button of the Startup and Recovery section. In "System failure" uncheck the option called "Automatically restart". Now see if you get an error message during reboots. Post it here.
3) Find the brand & model of your hard drive, and go to that manufacturer's website. Download the hard disk diagnostics appropriate for your model. Usually these will create a bootable floppy diskette & run the diagnostics from DOS. Some newer diagnostics can run from bootable optical disks, CDs and DVDs. Even though the CheckIt utilities (which I'm not familiar with) may have tested the drive, the diagnostics from the hard drive's manufacturer are generally the most effective.
4) Double-and-Triple-Check your Bios settings. If you are accidentally overclocking, you may be overstressing a component & it's leading to data corruption. If you are overclocking on purpose, return to stock speeds while troubleshooting.
--Also, make sure that the system's time and date are correct in the Bios, in case your motherboard's battery is ending it's life a wee bit early. A faulty Bios/CMOS battery can cause strange reboot issues. The nice thing about such batteries is that they only cost about $1.50 (USD) to replace.
5) On systems that are having trouble with power regulation at the motherboard/pci component level, I recommend disabling the "Spread Spectrum" setting (usually under Power Management).
6) Since USB hubs aren't expensive, try either borrowing a different powered hub, or buy another.
7) Don't overlook the obvious places for clues - your Event logs in the Windows XP Administrative Tools. Check that Disk Management reports the drive as "Healthy" and not "Dirty". If "Dirty", you can try a run of chkdsk with the /F switch (takes a long time, you can run it in Safe Mode if you'd like).
8) This is something I'm including just to be thorough (and for other readers with similiar system errors like yours) -- sometimes in a hurry, I've seen computer owners forget to replug in both power connectors to the motherboard when replacing a power supply - generally, it's the four-pin plug that is forgotten. I imagine you haven't forgotten that detail, but I've put it here just in case. [Yes, it leads to reboots when disconnected].
9) Can't hurt to use a can of compressed air to blow out dust that may have collected on the motherboard/cpu/heatsink/fan. Of course, follow the static precautions, and have the system unplugged during this.
10) Run thorough antimalware scans on the system also. They can corrupt files too!
11) Try a run of Windows File Protection (if you have an XP disk with SP2 on it handy). From the Start/Run box, type "sfc /scannow" (without the quotes).

If I had to guess, I'd take a longshot bet on a troublesome video driver (an error message would give us an answer on that). More likely, power trouble led to data corruption.

Sorry if this goes on a bit long
. . . Gary

[mevensen]

Originally Posted by GaryRouth 1) Start the system in Safe Mode. See if you have any problems with rebooting while running in Safe Mode for a somewhat extended period of time. Yes, your graphics won't look wonderful, but it can help with figuring whether this is a hardware or software problem.

I haven't tried Safe Mode yet, is it enough to let the system idle in safe mode to find a potential problem, or do I have to try to run software in that mode?

2) Change the setting for Windows XP that causes the system to automatically reboot when encountering a fatal error. If a driver or other conflict is causing the problem, you aren't able to read the error message - unless you turn this behavior off. To do so, click on Start, then right-click on My Computer, then select Properties, then Advanced, then click the Settings button of the Startup and Recovery section. In "System failure" uncheck the option called "Automatically restart". Now see if you get an error message during reboots. Post it here.

Did that. Today, I've only had two crashes in about 8 hours of uptime. One was a system freeze, requiring manual power off. No message.
The other was a crash with a blue screen saying "PAGE_FAULT_IN_NONPAGED_AREA, STOP 0x00000050". I tried to look this up in the MS knowledge base. the PAGE_FAULT error was said to correspond to an error in a Windows NT Server build, nothing for XP. The Stop code was not listed.
I hope not to see any more crashes, but if I get a meaningful error, I'll post it.

3) Find the brand & model of your hard drive, and go to that manufacturer's website. Download the hard disk diagnostics appropriate for your model. Usually these will create a bootable floppy diskette & run the diagnostics from DOS. Some newer diagnostics can run from bootable optical disks, CDs and DVDs. Even though the CheckIt utilities (which I'm not familiar with) may have tested the drive, the diagnostics from the hard drive's manufacturer are generally the most effective.

Tried this, but there was a problem with diagnostics program, apparently, it doesn't play well with Norton GoBack. I hesitate to Uninstall GoBack, because last time I tried it caused all sort of problems that took a while to fix.

4) Double-and-Triple-Check your Bios settings. If you are accidentally overclocking, you may be overstressing a component & it's leading to data corruption. If you are overclocking on purpose, return to stock speeds while troubleshooting. --Also, make sure that the system's time and date are correct in the Bios, in case your motherboard's battery is ending it's life a wee bit early. A faulty Bios/CMOS battery can cause strange reboot issues. The nice thing about such batteries is that they only cost about $1.50 (USD) to replace.

Date and time are correct. All other settings seem fine. I noticed that the graphics adapter was set to PCI instead of AGP, but I think this is a relatively new error, after I updated the BIOS at the end of this week.

5) On systems that are having trouble with power regulation at the motherboard/pci component level, I recommend disabling the "Spread Spectrum" setting (usually under Power Management).

I couldn't find any mention of this in the BIOS setup

6) Since USB hubs aren't expensive, try either borrowing a different powered hub, or buy another.

Have already done so.

7) Don't overlook the obvious places for clues - your Event logs in the Windows XP Administrative Tools. Check that Disk Management reports the drive as "Healthy" and not "Dirty". If "Dirty", you can try a run of chkdsk with the /F switch (takes a long time, you can run it in Safe Mode if you'd like).

Disk Management reports both HDDs as Healthy. I have run chkdsk on them recently.
The Event log notes some system errors, but no specific error code is repeated. None of the ones I looked up were present in the MS knowledge base.

8) This is something I'm including just to be thorough (and for other readers with similiar system errors like yours) -- sometimes in a hurry, I've seen computer owners forget to replug in both power connectors to the motherboard when replacing a power supply - generally, it's the four-pin plug that is forgotten. I imagine you haven't forgotten that detail, but I've put it here just in case. [Yes, it leads to reboots when disconnected].

Checked and confirmed.

9) Can't hurt to use a can of compressed air to blow out dust that may have collected on the motherboard/cpu/heatsink/fan. Of course, follow the static precautions, and have the system unplugged during this.

Already did that.

10) Run thorough antimalware scans on the system also. They can corrupt files too!

Scans clean.

11) Try a run of Windows File Protection (if you have an XP disk with SP2 on it handy). From the Start/Run box, type "sfc /scannow" (without the quotes).

I don't have this.

If I had to guess, I'd take a longshot bet on a troublesome video driver (an error message would give us an answer on that). More likely, power trouble led to data corruption. . . . Gary

I checked the device manager, and there's no alerts in there. I recently updated the video device driver (middle of this week) to the most recent available, uninstalling the previous. Same with the sound card driver.

[scomac]

I did a google search and came up with this on the 0x50 error:
http://kadaitcha.cx/0x50_error.html

I also found this on the PFN_LIST_CORRUPT:
http://forums.windowsforum.org/lofiv...hp/t22117.html

Scomac

[bailey]

check for dust build-up in the fins of the heatsink.

[GaryRouth]

Hi again

You've been pretty thorough, but there are a few things left to try.

As far as Safe Mode goes, if you leave it on for a long time, and give it something lengthy to do (like a Disk Defragmentation), and perhaps try a few programs (open & close your word processor, your imaging software, etc.) . . . and nothing breaks - that would tend to put more weight on a software error. If it crashes in Safe Mode, it tends to point to hardware.

Why not try one more memory tester (since I'm not familiar with CheckIt), just to see. Try a run of MemTest86 ( http://www.memtest.org ) -- it's true that a majority of the errors you're seeing are memory problems.

The fact that the hard disk diagnostics wouldn't run is puzzling. Norton's GoBack shouldn't be involved, since the diagnostics - when started from a bootable floppy or CD - usually start in DOS. Try the diagnostics again - if they generate an error message, post it here & we'll have a look.

You mention having trouble with both GoBack and SystemWorks. You have a lot of company in that regard. I don't see many Gateway systems, but for those that I do fix, the first thing I do is remove GoBack. Same goes for SystemWorks. Depending on what version of SystemWorks you have, the darn thing is known to install it's own rootkit (it's just not as well-known as the Sony copy-protection one). SysInternals has a tool that can try to find such an uninvited guest http://www.sysinternals.com/Utilitie...tRevealer.html --- I also highly recommend their Process Explorer utility.
_______________

And, of course, since everything ran fine for two years - and you still have original hardware inside the box (with the exception of the power supply), you can back up your data & run the Gateway Recovery disks. --Might call Gateway, too, just to make sure they didn't use proprietary wiring on their power connector on the motherboard, and that a generic supply can be used (just to cover all the bases).

Best of luck
. . . Gary

[mevensen]

Thanks for all your suggestions, I will try the memtest next, as I've been wondering the same.

As far as the Gateway products go, I had read that Gateway does not use proprietary parts, this was one of the reasons I went with them, since I had had issues with a Dell previously with obtaining replacement hardware. I can check again, though.

Actually, I've been running stable for most of the day now (knocks on wood), so maybe one of the myriad of things I've done in the last 24 hours worked (prays).

[mevensen]

So, I've been able to run for 1 1/2 days without a hitch, then I get all sorts of crashes and inability to boot, including the inability to get into the boot routine (don't even see the video card or BIOS screens).

So I run memtest. I had gotten it on your recommendation, but hadn't bothered to run it since things were apparently running smoothly over the last day.

It returns with >100,000 errors! Now, I'm not familiar with memtest, but I assume that's bad. My question is, how bad? After I post this, I'll be pulling RAM modules to see if I can identify a bad one. But, I'm (once again) puzzled how memory apparently addled with errors can run for over a day without a problem, and then hit me again with horrid crashes, etc. Does this make any sense?

BTW, I have no idea if this is coincidence, but after things were running well for that day, I used the SanDisk Cruzer Profile jump drive again. Shortly after using that, the crashes started again. It doesn't seem like using a jump drive should cause such problems, or even be responsible for revealing them, but I just wanted to mention it.

[Panama Red]

Any memory errors is bad. Test each module one at a time to determine which is defective. Chances are you were able to run fine for a few days because the defective areas of the ram weren't being accessed. I doubt the Cruzer had anything to do with the problem, just coincidence

[GaryRouth]

Curiouser and curiouser (from Alice in Wonderland).

Like Panama mentions, usually the USB device wouldn't be involved. Although - if other devices plugged into the same USB connector act up - you might have a faulty plug. I had a customer with an older KSEM board, that when any USB keyboard or mouse was plugged in, the computer wouldn't boot (wouldn't even get past POST). Without USB = everything was roses. Never a dull moment.

Most likely in your case, though, it's just a bad memory module, or the timings are off in the Bios.

Once this one's fixed, you'll finally get to relax & enjoy your Gateway again: that's a pretty capable box.
. . . Gary

[. . . and Happy Birthday Panama ]