Random Reboots on 3 day old rig - Windows 7 build 7000

I seem to remember reading sometime back about the NSK2480/Fusion having a grounding issue with the front panel. I have a Fusion box that sometimes crashes when plugging something into the front panel. I've only recently detected this pattern as it happens so infrequently. I haven't really worked at it yet.

So not an answer, but a question: Are you doing anything (plugging in a USB, touching the case, etc) when your system reboots?

Other than that, there are the usual suspects of RAM compatibility or the PSU.

do you have the system set to reboot upon a BSOD, if so disable it , if it happens again at least maybe you'll have a info from a bsod to go on

(my apologies if thats not an option within windows 7) no exposure to that OS yet.

Wow 61 degrees C? That's quite hot for a processor. I bet your CPU is overheating and the thermal shutdown in the BIOS is set to 60 degrees (in which it shuts off the system).

Ha you're talking about "debugging" and a beta OS on AVS? You might want to try Microsoft TechNet or something instead, LOL.

Though I'm not saying no one here can help you I'm just saying this doesn't seem like the usual thread for the HTPC forum on AVS...


No that's his system at full load (Orthos), it's not the usual temp or the temp he's necessarily getting when it reboots.

Though 61C is a bit hot for a stock clocked [45nm] E8400 (what is this passively cooled or something?) it's still well within it's acceptable temp I believe.

60c is not to hot for a processor at all

infact it shows his proc has been running at 60c for the last 10hours.

That 60c is at full load.

My borther's core i7 does a solid 75c when laoded with 8 thread for an extended period of time.

It's hard to know what the issue is.

How long is it between reboots? every day, twice a day, once ever 3 or 4 days?

If you are getting crash times often enough to do some trial and error work without having to wait 2 weeks to verify try this:
Unplug everything from the motherboard except essentials (on/off headers, video card, ect).

I second this. I am not familiar with Windows 7 but NT kernal in general has the auto reboot enabled by default. Disable the option and see what your BSOD is. If it is just a hard reboot from BIOS then it is most likely heat related. The stock E8400 fan (Which I had in my box) does not blow worth a damn. Get a nice after market fan like: http://www.newegg.com/Product/Produc...82E16835186134.

My system runs about 10c cooler than stock (non-overclocked) after adding this heatsink + overclocking the system by 25%.

Sure its not your power supply... this is a typical event when your PS is starting to go... It can either precede by a BSOD or just a simple reboot.

I just noticed he has a sub 400watt PS. That's pretty dinky. It is possible that the PS is not physically failing but failing to provide enough power for the system when under load which can cause all sorts of anomalies.

(Edit: I am guessing that was the PS that came "free" with the case.)

More likely to be a RAM problem. I would start there pulling out 2 gb of ram and see if the problem goes away then switch, etc. Or 1 stick at a time with memtest for several hours (overnite)

BB

I think a 400W PSU is more than enough for this system. I'd be surprised if the setup uses more that 200W peak. The PSU that came with the case is not shabby. I have this case. The PSU might still be defective.

I think it's probably the RAM. What happened with MemTest?

Also try removing the wireless card. I've had nothing but trouble with wireless and htpc. Wireless cards are especially troublesome, bad drivers, problems with wol, pauses due to scanning, etc. If you really need wireless, get a wireless client connected to ethernet port.

Also try removing all the cards.

Sometimes you just have to blame the motherboard.

Your power supply should be plenty, but it always could be bad.

Memtest would be a good idea, although the load testing you were doing frequently uncovers memory problems as fast or faster.

It sucks trying to errors that can't be reproduced at will.

Im running a 300 watt PS with a similar build with no issues less the tuner and extra HD..... So 400 is plenty. I believe you should be fine. You will only hit your max utilization at 100% load - your probably less then 50% at any time.. Your probably pulling less then 300 watts during normal use.

Either your PS is faulty, faulty card, faulty memory or faulty MB OR Bad Driver.

Under what condition has it rebooted? During playback? doing nothing? file transfer, etc?

Has it rebooted during the Memtest AT ALL? If so, then its probably Memory or MB.

Have you run Prime95 at all to burn in the system? If you get a reboot, lockup OR a failure then its memory or MB (could also be a setting in BIOS - make sure your using all defaults with any devices you dont need turned off) OR could be any of your expansion cards (Happauge)... Pull out Happauge rerun test - if reboots/fails, pull out 2 gigs of ram rerun - if reboots/fails, rerun with other set of memory - if reboots then its either MB or PS..

My gut tells me its your PS - go into your BIOS and verify the voltages to the components - anything with greater then a 3% (3% is the guideline but 10% is the number I go for) deviation from the norm could potentially cause instability - especially if its lower OR Driver issue..

Windows 7 has been unbelievably stable for me but I am running an NVidia setup - I would assume the Intel setup would be as stable or more then Nvidia (Nvidia is not known for having stable drivers).

Yeah it's time to bust out the troubleshooting skillz, lol. It will take a lot of your time though (I hate dealing with crap like this myself, end up spending hours on end and just having a headache at the end of it all ).

Your RAM seems like it's fine. You may want to run Memtest for 4-8hours though, just to be sure. (Edit: nm I see you've run it for like 10hrs already.)

You're going to have to try one thing at a time, remove your cards (network, TV) and stuff as stanglx said, try with none, then if you don't have the problem try adding them in one by one and re-testing.

Try another PSU (unfortunately if you don't have one handy and can't borrow one you may have to buy one to be sure). The unfortunate thing about hardware troubleshooting is that you have to have replacement components to test as well, which not everyone has available. As most have said 380W is certainly enough for your system and Antec makes fine supplies, but that doesn't eliminate the possiblity that the PSU is defective.

The other thing is, using a Beta OS is not the best thing to be testing on either. If you have Vista (or XP) you might want to switch to that for now, while you test the hardware out.

After all that if you still can't figure it out it may be the board. Intel mobos are generally known for being bulletproof, but there is always that chance. There have been so many times I've gotten a "new" motherboard that was crap right out of the box (though that's happened a lot less these days than it did in say the late 90s where you had like a 50/50 chance of getting a bad board, lol).

I can tell you from experience you may have a failing HD.
I've had this issue personally where the computer would crash with no video output and sure enough it was the hard drive.

Yep.. lockup could be this too... but a quick deep scandisk would surface any issues... PLUS you would also have some data corruption... I dont think its this as he is not getting any DR.WATSONS or specific application exceptions...

Install another version of Windows and see if you still get a random reboot. If you don't, report the problem to Microsoft as it's likely a Win7 issue.

Right now, there are just too many possible variables that might be causing the issue and the OS is an easy one to eliminate...

Neither was I
It was a partial new build that was driving me crazy because I thought for sure it was probably the HDMI driver. Then, the HD finally bit the dust....and so did another one soon after that while having the same issues. Then, I finally used a brand new disk and didn't have a lockup ever again

Also, note EVERYTHING was the same driver wise as I used disk images, even re-imaging the original disk that was in there before it died.

I'm not saying this IS the OP's problem, but definitely do a complete scan of your drive.

Yup!
Without proper diagnostic tools or pop-in replacement components (memory, hard drive, PSU, GPU, Sound card, optical drive, etc...) an OS reinstall with a proven (non-beta fish) OS would be the most straight forward approach. For all we know- it could be a motherboard problem... but when you have a new unproven system build and a beta OS there are just too many possibilities. Load Win XP Pro and go from there.

I would also suspect RAM first. Have you done any tweaking in the BIOS? I ask because the RAM you purchased is spec'd at ddr2-1066 but according the Intel's site your board only supports ddr2-800. The board will automagically clock the RAM to 800MHz unless you've set the timings manually. Either way I'd reset to BIOS defaults before doing anything else.