Windows 7 Random Restarting With No BSOD

[Dragomar]

Hi there, I've been all over trying to solve this issue and nobody seems to have any answers. Hopefully you all will be able to help =P

My System:

EVGA SR-2
2x Xeon E5620
2x Cooler Master Hyper 212 EVO (both in Push-Pull)
24GB Corsair XMS3 1333mhz RAM
Zotac GTX 670 4GB
OS HDD: Western Digital 2TB Black
8x additional case fans
Corsair AX1200 PSU

I built it myself and it's my first custom computer that I built with my own two hands (I've had other high performance systems from the likes of IBUYPOWER, etc.)
It was finished late December and has been running perfectly for months, untill the 14th of this month when it restarted out of the clear blue sky with no BSOD.

Ever since it's been happening very intermittently, and I cannot replicate what causes it. The system has restarted with no BSOD (to verify that there indeed was no BSOD I disabled Automatic restart as so many sites point to, still no BSOD. It's just working one minute and then black screen and reboot another) It'll happen while gaming, while browsing the internet and listening to minus, while using VLC... It's really the only stain on this otherwise perfect-for-4-months computer. It isn't heat (http://i159.photobucket.com/albums/...AE3D38-28463-000005F430981A3E_zps82cb9714.jpg)
as the processors run at 30C or below idling and around 55c at full load, and the random restarting problem doesn't happen any more often at load then when it is idle.

Things I've tried to attempt to fix the problem/find out it's cause, all to no avail.

Plugged computer directly to wall, as I've heard power strips can cause the problem.
Replaced OS HDD's SATA cable
Memtest86'ed, it passed.
Western Digital's Data Lifeguard Extended Test on main HDD, passed.
Uninstalled/reinstalled latest nVidia drivers

Here are all of the times it's restarted out of nowhere, with the exception of a few minutes ago, 5/28/2013, 1:28:17PM. I realize these are not the actual times of restarting but they're a few minutes after when the system reboots.
(http://i159.photobucket.com/albums/t148/Hogdriver155/flawless_zps42c02db6.png)

I did notice this error (http://i159.photobucket.com/albums/t148/Hogdriver155/cause_zpsb2524930.png) very near to the Kernel Power restart thing.

The second part of it references my CD drive... No clue what it means, however.

Thanks for your help, really appreciated...

 

[Digerati]

You can't rule out heat by CPU temps alone. There are many heat sensitive devices on the motherboard. One problem with aftermarket coolers that don't fire (blow air) in the conventional manner down onto the OEM heatsink (besides voiding the warranty on boxed CPUs that come packaged with a supplied OEM cooler) is they don't provide the expected air flow pattern motherboard designers expect. So unconventional aftermarket coolers (like your side firing cooler) don't always provide enough help to all the heat sensitive and heat generating devices that typically surround the CPU socket. So I would blast a desk fan in to the open case and see if it reboot.

Depending on your location on Earth, summer is coming and warmer weather can greatly affect ambient (room) temperatures too - which of course, affects computer temps.

Did you apply a nice thin, proper layer of TIM (thermal interface material) for the CPU cooler?

A failing power supply is a common cause of sudden freezes, shutdowns, and reboots. I would swap in a known good PSU.

While Memtest is good, no software based memory tester is conclusive. Memory can test good and still be bad - or simply refuse to play with other sticks. You might try running with one stick at a time.
And if overclocking, stop - until this is resolved.

 

[Dragomar]

Wow... how very right you are. CPU temps were fine, but somewhere along the line, EVGA Precision X (program that allows you to control onboard GPU fan speeds according to load and temperature, in addition to overclocking the memory and core, if you didnt know, if you did, sorry to be patronising) got disabled, and didn't start on windows bootup, so my GPU was overheating... I expect this may have been the cause. It was in my mind, previously, that GPUs could only throttle themselves back, and not restart the entire system, if they overheat? Apparently not!? Now things are much better in GTX 670 land that my original agressive fan curve has been implemented. Hopefully this will fix my problems. I also found that my BIOS was outdated... by 2 years. So I updated that as well.

I'll post back if I have any other restarts...

 

[Digerati]

if you did, sorry to be patronising)

No one can assume they know the areas of knowledge and levels of expertise of the original poster or of those helping. Therefore, it is not patronizing but being thorough - so EVERYONE can follow, including those just reading along now, and those reading in the future who are searching for help for their similar problems. If someone on these forums gets mad, or takes it personal when you tell them something they already know - that's their maturity (or lack of it) problem, not yours!

that GPUs could only throttle themselves back, and not restart the entire system

They can't restart the system (not by design or coding, that is). But a failing or stressed graphics solution can put a "condition" on the power, interface, or OS that causes a fault in one of those or an associated circuit, which in turn, causes the system to halt, or reboot. A failed graphics solution can also prevent a system from booting.

I also found that my BIOS was outdated... by 2 years. So I updated that as well.

I am glad the update appears to be working but for future reference, it is not normally a good idea to update the BIOS just because a new version is out there. Most BIOS updates are not fixes, but rather simply add support for new RAM or new CPUs that came out since the last update or board release, and have nothing to do with your current system.

While the process of flashing the BIOS has greatly improved in recent years to prevent any problems, the process is still very risky as an update failure can, and sometimes does end up killing the motherboard - permanently. An untimely power outage in the middle of the flash, for example, may ruin your day - and checking account.

I only update the BIOS on my systems when the list of changes says the update (1) addresses a problem I am experiencing, (2) current BIOS is corrupt (rare) or (3) the update addresses a security issue (also rare).