Always restarts in middle of boot-up

My computer has just developed an annoying little quirk: it spontaneously restarts in the middle of Windows start-up. I have Windows XP Pro SP2, and every time it gets to the part of booting up where the little gray bars fill up across the bottom of the screen (right before the Windows logo is displayed on a black background), it will all of a sudden just restart and go back to the BIOS screen. So basically, it can never get past this point, and fails to give me any sort of message or information as to what exactly the problem is.

Here's a little video in case it's not clear what part of boot-up I'm talking about.

Even stranger is how the problem started: I was just doing typical everyday work on the computer when everything froze, the mouse cursor disappeared, and it would not respond to any input. I had to do a hard restart, and that's when the boot-up problem began.

This is not the first time this boot-up restarting thing has happened either. The first time I experienced it was a week ago and I ended up doing a clean re-install of Windows to fix it. But apparently that wasn't good enough, because it has resurfaced. So rather than re-install Windows again, I'd like to find out once and for all what the cause is.

Here is what I've already done to try to troubleshoot this:

1. I tried to start up using "Last Known Good Configuration ", but this didn't fix anything.

2. I tried to start up in Safe Mode, but again, the same problem.

3. I tried to enable Boot Logging so I could see at what point exactly the problem was happening, but it never seems to produce the boot log file...or at least I can't ever find it. After enabling Boot Logging, and the boot-up fails, I enter the Recovery Console and try to find the "C:\WINDOWS\Ntbtlog.txt" file, but it is never there.

4. Next I started to remove pieces of hardware from my system one-by-one, trying to boot up after each one to see if it fixed anything. Now the only things plugged into my motherboard are my video card and one hard drive. Still, same problem.

5. Then I decided to get drastic and starting disabling start-up Services using the Recovery Console. I wrote down all the services marked as "SERVICE_BOOT_START ", meaning they get loaded during the boot phase. I disabled them one at a time and discovered that the problem was still happening when all of them were disabled, except for the "ATAPI" service. The catch there is that when I disabled "ATAPI" I got a Stop message during boot-up that said it could not continue because it couldn't find the "C:\WINDOWS\System32\Drivers\Ntfs.sys" file. So I can't say whether or not the "ATAPI" service is the one at fault for this spontaneous restarting business, because I have no way of checking for a successful boot-up when it is disabled.

Anyway, I've pretty much reached the end of my rope now, and I'm getting sick of re-installing Windows XP to fix this problem every time, since the true cause is something deeper than that. Any suggestions?

Here are my system specs (keep in mind only the video card and Maxtor 5T06H6 are hooked up at this point):
OS: Windows XP Pro SP2
Motherboard: ABIT KG7-RAID
CPU: AMD AthlonXP 1800+ (not overclocked)
Memory: 2x256MB Crucial CT3272Y265
Video Card: Sapphire ATI Radeon 9550SE
Sound Card: Creative Labs SoundBlaster Audigy Platinum
Network Cards: 3Com 3C905CX-TXM, D-Link DWL-520+
Hard Drives: Maxtor 5T06H6, Maxtor 6Y120P0, Maxtor 6Y200P0
CD Drives: Toshiba SD-M1502, Ricoh RW7200A

Thanks for reading all this

 

MOV file

Your video link doesn't work.

 

You may have a defective memory stick. Pull one and test each separately or else get the Ram Tester and create a diskette to run a test from.

Aside from that you may want to get the HD diagnostic tool from Maxtor and test the boot drive. I don't think that's the problem but you have to eliminate it.

Then of course that leaves the video card, power supply and MOBO. Since it doesn't boot to Safe mode either, the video card isn't likely at fault but still needs to be substituted if you can.

Good luck.

 

Update:

Ok, I have tried out all the things people have suggested so far: I swapped memory modules, did a memory tester (both MemTest86 and Windows Memory Diagnostic), checked my hard drive using both "chkdsk /r" and Maxtor PowerMax Diagnostics, cleared all my CMOS settings back to fail-safe defaults, and tried replacing my viaide.sys driver with the lastest version. Unfortunately, none of these things fixed the problem. As a side note, yes, it does let me boot up with a boot floppy, a boot CD, and the XP Recovery Console.

Now here's something new that I've discovered: I was trying to focus on booting up in Safe Mode, since Safe Mode is kind enough to actually print out the names of the drivers as it loads them. First, I disabled the "atapi.sys" driver because I knew that would cause a Stop Message, and I wanted to see which drivers got loaded before that Stop Message. Here is exactly what Safe Mode printed to the screen:

Code:

multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\ntoskrnl.exe
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\hal.dll
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\KDCOM.DLL
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\BOOTVID.DLL
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\config\system
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\c_1252.nls
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\c_437.nls
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\l_intl.nls
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\FONTS\vgaoem.fon
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\AppPatch\drvmain.sdb
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\ACPI.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\pci.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\isapnp.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\ohci1394.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\1394BUS.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\WMILIB.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\viaide.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\PCIIDEX.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\MountMgr.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\ftdisk.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\dmload.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\PartMgr.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\VolSnap.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\hpt3xx.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\SCSIPORT.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\disk.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\CLASSPNP.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\fltMgr.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\hptpro.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\KSecDD.sys
[B]<Stop Message happens right here>[/B]

Then, I re-enabled "atapi.sys" so I could see how much farther it could get in loading the drivers, and for some reason it got less far. This is what happened:

Code:

multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\ntoskrnl.exe
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\hal.dll
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\KDCOM.DLL
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\BOOTVID.DLL
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\config\system
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\c_1252.nls
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\c_437.nls
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\system32\l_intl.nls
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\FONTS\vgaoem.fon
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\AppPatch\drvmain.sdb
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\ACPI.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\pci.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\isapnp.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\ohci1394.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\1394BUS.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\WMILIB.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\viaide.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\PCIIDEX.SYS
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\MountMgr.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\ftdisk.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\dmload.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\PartMgr.sys
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS\Drivers\VolSnap.sys
[B]<Restart happens right here!!>[/B]

So, as you can see, when atapi.sys is enabled, the restart happens right at the point where hpt3xx.sys would get loaded, yet if atapi.sys is disabled it can successfully get through hpt3xx.sys. Does that make any sense to anybody?? Because I am at a total loss here, folks.

Your thoughts?

 

You have done a very good job of presenting the problem. Hang on, I'm calling in the big dogs for you!

Johanna

 

It seems pretty clear that you have a driver that is touching hardware in a bad way to force it to reset like that. The safe mode list is a good indicator of who is loaded, but not of who caused the crash. the ATAPI driver you were messing with is the IDE hard drive driver.

The next step in the troubleshooting process is to isolate if you are experiencing a software fault, or a hardware problem (bus reset or whatever) that is triggering the event.

My recommendation is to live debug the machine. You would need a second machine and a serial null modem cable. This will allow you to see if the phenomenon is occuring because of an exception or not, since the debugger would catch the fault and allow us to interrogate the software. If this problem is a hardware reset, the debugger would be useless, and troubleshooting would lead around swapping hardware in and out, mainboard, cpu, ram etc..


As an aside, the only 'unusual' thing i see is the firewire drivers, which i would assume are optional in this barebones config you are currently running in. It may be worth nixxing those as a quick test. ohci1394.sys 1394BUS.SYS


If you have a second machine and want to pursue a debug, i can post some instructions for the setup.

 

Jugdish,
There's one big WOOF! There may be other responses coming, too. Joe, you can take my vaccuum cleaner apart anytime...LOL

Johanna

 

One response would be to update drivers. Drivers seem to be at the bottom of a lot of SP2 problems. So that's one action to take in the meantime.

Regards - Charles

 

Capacitors have been ruled out already?
Second thought would be heat or bad/corrupted driver.

 

Hehehe ......... That is where your problems could have begun.

Tough board to set up. Known problems include conflict between the Highpoint controller and your IDE controllers. Any drives on the raid headers may have to be set up with nothing hooked up to IDE 1 & 2. No picnic here - involves pulling cables at each step of an installation because worst case scenario precludes you from having a CDROM drive hooked up at the same time as your HDD's on 3 & 4. Once you have an OS on any drive running from 3 or 4, you are home free and can use 1 & 2 provided the correct drivers are installed. Pain to set up but nice board otherwise despite the CPU limitations.

My guess is your clean install is frought with driver problems - Specifically your Highpoint drivers and quite possibly others.

Couple of suggestions -
Use BIOS Kg7ds
Use Highpoint 2.34 drivers
Assuming success with an installation, load only the AGP mini-port drivers and filter using AMD DRVPK 1.30 - reference your manual (read more about it) but download this stuff from ABIT.

 

Hpt3xx.sys is a driver for a Highpoint controller. Wonder if you use that? If not, you could try disabling it in the DM?

 

Update:

OK, as far as BIOS versions, I had flashed my BIOS to version DM a few months back, which I got from www.abit.com.tw. But here I'm a bit confused because both versions DM and DS have a compile date of 3/6/2003. Both include the latest Highpoint BIOS, version 2.34, so I don't see any difference between DM and DS. Honestly, though, I don't think BIOS is the issue here.

Now what I've tried is getting the latest Highpoint drivers (hpt3xx.sys and hptpro.sys) from www.highpoint-tech.com, which are version 2.351. I was hopeful after getting this latest hpt3xx.sys file, since this was the file (supposedly) that the system seems to be restarting on, but unfortunately this new driver did not resolve the issue.

Furthermore, I actually DISABLED the onboard Highpoint controller through my system BIOS, so now the Highpoint BIOS is completely skipped over during boot-up (I don't need it enabled right now anyway, since I've just got one HD on IDE1 and one CD-ROM on IDE2). Even this did not fix things.

JoeHobart, I think this is the route I would like to take at this point. I am absolutely determined to get to the bottom of this thing. Please post your instructions for doing a live debug, as I have no experience doing anything like that.

Thanks for all your help, guys!

 

re: my recommendation on your bios bin - feel free to disregard it, however, I didn't recommend without a reason.

re: my recommendation on Highpoint drivers and your comments in your second paragraph - careful - this from the Highpoint website -

Your Highpoint drivers must match your Highpoint BIOS.

re: hpt3xx.sys - disabling your RAID controller in the BIOS won't stop XP from trying to load your (possibly corrupt) drivers.

You should also check out your memory. Default voltage on this board is 2.65 and you may need to adjust it downward if you are using 2.5V memory modules. Three other simple things you can do to achieve memory stability on this board are set the "Control CPU P/N Value" to "4" , "04 ", "0D" in the BIOS, increase the CPU Slew rate to 7 and lastly you may need to increase the FSB to 136 or 138MHz to achieve stability.

Having said that, good luck .............



edited afterthought: sorry it comes abit (pun intended) too late because I think you will need to do another installation but on a lot of these older ABIT Boards moving memory to different DIMM slots ie: 2&3 or 1&3 vs standard 1&2 resolved the symptoms you referenced earlier. Regardless, think Joe has a good concept there and I'm anxious to learn just how this is done.

 

Your first problem, a hard lock-up, sounds like you could have a hardware problem. After the hard reset, your OS kept rebooting most likely because it found a software problem caused by the hypothetical hardware problem and you have the default set in XP where it just keeps rebooting and rebooting whenever it encounters a serious problem.

Memory is always a good choice of culprit but apparently you've already troubleshot that.

There is the possibility that your PSU is not being nice. Either putting out insufficient juice or doing so in a very rude manner. There's also the possibility that the power going into your box isn't as clean as it should be but I can't recommend any testing methodology for that. I do use a PSU tester that tests all 3 rails and lights up like a pretty green Christmas tree if all is okay. Do you have a spare PSU lying around? What is the output (W) and brand of your PSU?

Just reading about Rockster's talk about skew settings made me a bit dizzy. It seems to me that if a board's default BIOS settings weren't good enough for successful running that I wouldn't think the board would sell very well.
And RAID can be a real wildcard when you have a box that's misbehaving.

Gary

 

sidebar - Chiles4 - 100% correct, thats why it was taken off the market so quickly.

 

My 2 cents.
I looks to be a graphics problem to me?
Because the machine reboots just as the PC switches from VGA mode to SVGA mode. Could be a coincidence but...?

If you can boot I would suggest uninstalling any ATA drivers and try with the generic ones that are in XP. They're 3+ years old but may prove the fault.
If you can get up and running try updating to the latest ATI Radeon drivers. If like Nvidia, Radeon have reference drivers, try those. I only use the Nvidia reference drivers rather than the board manufacturers (Leadtek) ones.

Lastly, try another video card.

HTH?

*EDIT* Actually, reading your post again, it may STILL be in VGA mode till the screen goes momentarily black AFTER Windows is displayed. So my VGA/SVGA switch theory mau be a few seconds premature???

 

Combining/reinforcing what both Paul and Chiles4 have said - I've seen ATI Radeon AIW cards cause sudden shutdown or instant freeze/lockup because of the power they require. Don't know what your card pulls but its a lot more video card than the AIW's I had problems with.

 

I would clearly point out we have several resident hardware experts, who have given you some very good information. I do not see any problem in working the phenomenon in parellel, as determining if the fault is software generated will be helpful in determining the course of troubleshooting past this point.


Heres the basic description of how to kernel debug:
1) Get the tools: http://www.microsoft.com/whdc/devtools/debugging/installx86.mspx
** use the Debugger.CHM help file for detailed instructions on these points under the 'installation and setup for a live kernel debug'
2) Install them on a second machine
3) Obtain a serial null modem cable to connect the two machines together via serial port
4) Make a boot floppy KB article
5) edit the boot.ini on the floppy to include /debugport and /baudrate setting appropriate
6) run the KD debugger on the second machine and verify that you have connectivity by rebooting the target machine (you'll get chatter in the debugger)
7) setup a symbol path pointing to the internet symbol server SRV*c:\websymbols*http://msdl.microsoft.com/download/symbols
8) do it

If the problem is software generated, it should drop into the debugger rather than reboot. If not, then most likely you have a hardware problem causing the phenomenon (although that is not an ABSOLUTE certainty)

 

Thanks guys for all these wonderful suggestions!
I'm going to try them out, but I probably won't have time until this weekend. So expect a follow-up diagnosis from me sometime this weekend at the earliest.

So glad I've finally found a Windows support forum with some truly helpful + knowledgable people on it!

 

Jugdish,
Please post back after you've had some time to work on the machine. This BBS is a lot of fun because of the diversity of the members, and their willingness to pool their experience and knowledge.

Thanks! (WOOF!!) to all you big dogs for stopping by!
Johanna

PS Rockster- I'll PM you later. Got a question for ya!