Random reboots - looking for advice

[fogus]

Hey, everyone. Happy winter and such. (I just escaped some near-frostbite today--yikes! Had fun skiing though.)

I'm back for some advice.

My PC is randomly rebooting and I think it has to do with hardware but I'm not sure.

Tech specs:
* PSU: Corsair Professional Series AX 750 Watt ATX/EPS Modular 80 PLUS Gold (AX750)
* Mobo: ASUS P8Z77-V LX LGA 1155 Intel Z77 HDMI SATA 6Gb/s USB 3.0 ATX Intel Motherboard
* CPU: Intel Core i7-3770S Quad-Core Processor 3.1 GHz 8 MB Cache LGA 1155 - BX80637I73770S
* Graphics: XFX AMD Radeon HD 6870 3D-Ready Graphics Card (HD-687A-ZHFC)
* RAM: 32G @ Crucial CT2KIT102464BD160B 16GB kit (8GBx2), 240-pin DIMM Upgrade for a ASUS P8Z77-V
* HDDs: 2 @ Intel 128G SSDs (non-RAID) and one 2TB WD.
* Cooling: custom external water

Everything except for the graphics card (maybe?) is still in warranty. The PC was built in Oct 2012.

This has happened about four times so far. Typically, after leaving the PC mostly idle (think heavy download/upload traffic but no graphics running) I find the computer waiting for me to enter a pre-boot password. Uptime has been about 24 hours on average since then.

Tried so far:
* Checked for crash dump and system/app/security error logs (nothing shows up except the subsequent "the last reboot was not expected". before the period of outage, the PC seems normal)
* Checked internal cabling. Replaced one SATA cable.
* Checked temps... seem normal (37C CPU).
* Checked voltages... seem normal to me (my vcore is about 0.924 to 1.032. Is that low? I haven't overclocked in years)
* Checked HDD health... seem normal, but I didn't go too deep.
* memtest.org for 36 hours straight (while on my ski trip). No reboots happened during this time. No errors either in my beautiful 32 gigs. Hmmm
* Tested my UPS. It works. I get about 90min+ with my monitors off (which is how I leave them when I sleep). The UPS alarms if it looses power and someone would have heard it if it did. It has woken me up in other power outages. The UPS says the voltage and frequency of our power is good (119V/60.0Hz with 118-119V output).

Trying tonight:
* Voltage logging enabled (I hope the stupid ASUS software stores those persistently)
* Logging Procs and Services with powershell to all three connected drives simultaneously. Hopefully when it crashes again, I'll have the last moment (and possibly also know if a drive decided to hang before the others)

Will try next if I don't get useful data from this next crash:
* Unplug everything unnecessary and run (e.g., no peripherals, one monitor, one ram stick)
* RMA the motherboard (assuming I can't prove the PSU is a problem)

The mobo would suck to replace, but it's my highest priority because the PSU has a 7 year warranty and the mobo has a 3 year and will be up this Oct.

Anything I'm missing? Am I doing this in a good way? Am I failing to think of something?

[FlyingPenguin]

First, make sure it's not the OS crashing. By default Windows restarts on a crash. If you turn off the auto-restart, then instead you will get a BSOD which will give you some useful info, and confirm that it's an OS issue (which may still be caused by hardware like bad RAM, but you would know it wasn't just powering off). To disable auto restart: Right-click My Computer -> Properties -> Advanced System Settings -> Advanced tab -> Startup & Recovery Settings button -> under System Failure uncheck "Automatically Restart".

If you get a BSOD the next time it happens, then do a Google search on the error code, which may or may not give you some clues. RAM would be a major suspect. I would re-seat the RAM (remove & re-insert) to clean any crud off the contacts. Some contact cleaner wouldn't hurt either.

Try running MEMTEST86+ all night and see if you get any errors.

If you DON'T get a BSOD next time, then I would suspect it's the PSU, but there's really no way to test that without replacing it.

[Nuby1Canuby]

My old Q6600 machine has been randomly rebooting for the last month.
The screen goes blank and you usually have to power if off and on with the power supply switch to get it to reboot.
It seems to occur more often when it first starts up or you are sitting at the desktop or browsing the net. It seems fine during heavy gaming.
I swapped the 4870 with the 6950 and it seemed fine for about a week but now it is worse then ever.
I haven't tried putting it on my UPS yet because 500 watts isn't strong enough for both PCs. I'm also avoiding swapping my Antec 650 watt PSU with the OCZ 500 watt until I have to.
The machine is about 6 years old and the power supply and video card are about 4 years old. The SSD is less then a year old.
It rebooted during pass 2 of the windows memory test so I can't see the error.
I reseated the memory and so far so good. I'll remove half the memory modules if it does it again. (I hate mid tower cases, the 11 inch 6950 and the CPU snorkel are in the way of the RAM)
It passed the Windows Memory Diagnostic this time so I think I'll button it up and see how it is over the next few days.

[FlyingPenguin]

Something I forgot to mention, and it may apply in your case Nuby: check the CMOS battery. They only last 4 - 6 years. A failing battery will cause all kinds of wacky problems, including random reboots.

Put a digital volt-meter on it and and if it's below 3 volts it's no good. They're so cheap it's worth just replacing it if you don't have a digital meter. It's a standard CR2032 lithium disc battery, available at any office supply store, or hardware store. It's also used in automobile remote key fobs so they're available everywhere.

[Nuby1Canuby]

Thanks, I'll check it with my volt-meter.