Computer suddenly started reseting randomly

[n00b]

Yesterday my computer started randomly reseting. Sometimes it would stay up for over an hour, sometimes it would only stay up for 10 minutes before reseting.

The day before yesterday my computer was running fine. I opened it up to clean all the dust out (too lazy to buy filters :b) of the whole case (even took CPU fan off heat sink and removed the dust that was on the heat sink). The only other changes that were made was the fan RPM sensor from my power supply was moved to a fan head on my MB that compatable with RPM sensing (I have a Soyo Dragon+, so the BIOS only checks RPMs from CPUFAN1 and CHAFAN1), the temperature probe from my Powersupply was moved to a different location, and some of my power/fan/ATA rounded cables were moved around so they wouldn't block airflow as much.

My powersupply is a 465W ENERMAX Whisper
Soyo Dragon+ with an XP1500+
5 case fans (3 in 2 out) besides powersupply so I have decent air flow, and a TI-V77 Dr.Thermal CPU cooler
My temps are pretty low -- BIOS says case temps are mid-upper 20Cs, CPU is usually low-mid 30Cs
1 CDRW
1 DVDRom
1 Floppy
1 7200 RPM HDD
Geforce 3 ti 200
Sound card
Chenming (SP :b) 601 case (same as the Antecs/Cheiftec/Alienwares etc)

Everything is less than 1 year old except the DVD ROM drive and the sound card.

Anyone know what the problem could be?
Too much stuff for my PSU (havent added anything to comp for 6 months)?
Dead PSU (Enermax's were supposed to last for a while I thought)?
Something else?

 

[Toejam31]

Power Supply issues are usually related to system lockups or sudden shutdowns ... not random rebooting.

Rebooting is generally either due to heat issues, a corrupted or incompatible file, or a driver conflict.

I would undo all the changes that were made with the fan sensors, and see if that helps. If not, you may have done some damage to the thermal pad or gel between the heat sink and the processor die while wriggling the HSF around when removing the fan, and you might need to reapply the compound. You can't trust a mainboard CPU socket probe to give you accurate temperature readings. In fact, the temperatures you are describing for your system that are being displayed in the BIOS are unusually low for a processor of this type.

For example, I've got a system with seven case fans, and I keep the ambient room temperature at around 66F, and still, my system temperature is not as low as yours, even with an aluminum case and a P4 processor.

You never mentioned the operating system version you are running. If you have recently updated a file or driver, this could be the problem. If you are running Win2K or WinXP, the Administrative Tools Event Viewer Logs could give you some additional information about the problem. With these operating systems, you should also be able to discover the source of the problem in two ways:

1.) My Computer. Right-click. Properties, Advanced, Startup and Recovery, Settings. Under System failure, clear the checkbox for automatically restart. Also set the Write debugging information to "none".

If the problem is due to a file or driver, this should force a Blue Screen that contains information about the error. Please document ALL the information carefully.

2.) A utility like <A HREF="http://msdn.microsoft.com/library/default.asp?url=/library/en-us/ddtools/hh/ddtools/dv_9rw3.asp" target="_new">Driver Verifier</A> can help you discover the source of the problem.

I don't believe your PSU is overloaded, not with a 465W Enermax; nor is it dead. If it was fried, you wouldn't be booting at all.

Toejam31

<font color=red>First Rig:</font color=red> <A HREF="http://www.anandtech.com/mysystemrig.html?rigid=17935" target="_new"><font color=green>Toejam31's Devastating Dalek Destroyer</font color=green></A>
<font color=red>Second Rig:</font color=red> <A HREF="http://www.anandtech.com/mysystemrig.html?rigid=15942" target="_new"><font color=green>Toey's Dynamite DDR Duron</font color=green></A>
________________________________________

<A HREF="http://www.btvillarin.com/phpBB/index.php" target="_new"><b><font color=purple>BTVILLARIN.com</font color=purple></b></A> - <i><font color=orange>A better place to be</font color=orange></i>.

 

[n00b]

I use Win2k Pro - Here is the error that came up when the comp crashed:

*** STOP: 0x000000BE (0xBA6DC423, 0x000D3060, 0xBA349B58, 0x0000000F)
*** Address BA6DC423 base at BA6BC000, DataStamp 3d6bb2c3 - nv4_disp.dll
Begining dump of physical memory
Physical memory dummp complete. Contact your system administrator or
technical group.

I'm guessing the problem is with the video drivers. I did upgrade my detonator drivers recently (to 40.41), so this may have caused the problem. Should I try to find the older detonator drivers and reinstall those?

 

[Toejam31]

You are definitely looking at an error that is due to the video card driver.

You should boot up in <A HREF="http://support.microsoft.com/default.aspx?scid=KB;EN-US;Q202485&" target="_new">Safe Mode</A>, and try to replace the driver with an older version, preferably one that is not a leaked driver still in beta.

There have been more than a few problems associated with the beta release of the 40.41 drivers.

I'd suggest the 30.82's, as a compromise, which you can still pick up from <A HREF="http://www.nvidia.com/view.asp?IO=winxp-2k_30.82" target="_new">nVidia</A>. Or you can simply install the driver that was previously used by the system.

Note: Never use the Setup.exe installer in a driver package from nVidia when upgrading to a newer driver set. The possibility of file corruption is too high. Instead, right-click, drag and drop, and extact the driver files into a newly created folder. Then upgrade the drivers by right-clicking on the Display Adapter in the Device Manager, going to Properties, and the Driver tab, and navigating to the new folder which contains the extracted files.

You'll save yourself some grief using this method.

Afterwards, disable the nVidia Driver Helper Service, in Control Panel\Administrative Tools\Services. This is unnecessary, and can interfere with the system shutdown.

You are also going to need to search for memory.dmp files, which have been getting, well, dumped on the hard drive each time the reboot occurred. These can be very large, if you had not previously adjusted the Write Debugging settings in Startup and Recovery.

Note: In your situation, it may be necessary to remove the video card from the Device Manager, and load a Standard VGA adapter with the Add/Remove Hardware wizard. Afterwards, you can navigate to the folder that contains the driver files, and and upgrade the VGA adapter in the Device Manager. Ths is an alternate method that is sometimes necessary in Win2K.

Good luck!

Toejam31

<font color=red>First Rig:</font color=red> <A HREF="http://www.anandtech.com/mysystemrig.html?rigid=17935" target="_new"><font color=green>Toejam31's Devastating Dalek Destroyer</font color=green></A>
<font color=red>Second Rig:</font color=red> <A HREF="http://www.anandtech.com/mysystemrig.html?rigid=15942" target="_new"><font color=green>Toey's Dynamite DDR Duron</font color=green></A>
________________________________________

<A HREF="http://www.btvillarin.com/phpBB/index.php" target="_new"><b><font color=purple>BTVILLARIN.com</font color=purple></b></A> - <i><font color=orange>A better place to be</font color=orange></i>.

 

[n00b]

I installed the detonator 30 drivers soon after my last post and rebooted to see if the problem ws fixed but it crashed again, and that time there was a different error.

Had some 0x numbers like the first error, but the main thing was:

MEMORY_MANAGEMENT

instead of the graphics driver like the first time.
Tried a few more times and got another error with:

IRQL_NOT_LESS_OR_EQUAL

along with 0x's like the first one. I decided it wasnt the GFX drivers, but probably the RAM. I had just installed a new stick the day before the problems started, but I had no problems that day (even ran a few programs that used almost all of my RAM) - they didn't start till after my dust cleaning so I didn't think that it was the RAM.

Well I switched switched DIMM slots with new/old RAM a few times and got different errors each time. One set up I would get an error while windows was starting up:

STOP: c0000218 {Registry File Failure}
THe registry cannot load the hive (file):
\SystemRoot\System32\Config\SOFTWARE
or its log or alternate.
It is corrupt, absent, or not writable.

Switched the RAM around again and windows would load to the desktop but then would not be able to load some of the programs/processes it loads at start up (different ones each time) and then windows could freeze or crash.

I removed the new RAM stick and left the 2 old ones in, and now my comp is working fine again and not crashing.

Must be a bad stick of RAM. Funny it's supposed to be the same brand as the other 2 sticks of RAM, but it looks a bit different and doesnt have the brandname logo on it like the old ones do. Last time I order from that site