Random Reboots - Help Plz

[xipetotec]

Hey guys, I've gotten so much help on this forum in the past, I'm hoping maybe you guys can help me again. This just started a few days ago, but I'm getting random reboots and can't figure out what the problem is. I've had a few BSOD's with it, but I think I found the problem with something having come unplugged from my Mobo, once I plugged it back in, the BSOD's stopped, but I'm still getting restarts and sometimes freezes when playing a game. I haven't had any problems booting up initially and logging in, everything is smooth, but then sometimes after 5 min, sometimes it takes 2 hours or so, it just reboots.

PC Specs:

Mobo: GB GA990FXA-UD3
CPU: AMD Phenom II 1090t BE
GPU: GB GeForce GTX 570
RAM: 2x4GB G.Skill Sniper 1600
HDD: 1x 64GB Crucial M4, 1x 2TB Hitachi Deskstar 7k3000
PSU: Corsair TX650
OS: Win7 Enterprise 64-bit

My build log is here if you want more info on what my OC info is. This PC is only about 8 months old.

What I've checked so far - temps were my first suspect since my system is OC'd, but I tested both CPU temps and GPU temps both idle and under stress, and temps were manageable. Today, I figured I would test the RAM, so I ran Memtest all afternoon while I was at work. When I got home, it made it through 4 passes (about 2 1/2 hours) and was frozen, but showed no errors, I had to reset to get it back on. I've tried running MSE and Malwarebytes to make sure I didn't have some kind of virus (they weren't installed b4 I started having this problem) but trying to run them several times and have always gotten a reboot before the scan finishes.

I recently set it not to reboot automatically, so hopefully the next time it does, I'll have an error code I can post to you guys and maybe that will help narrow down the problem. After doing this, I went to check my Event Viewer to see if any errors were popping up in there, and I see two consistent errors that pop up about every 1 hr 9 min:

Event filter with query "SELECT * FROM __InstanceModificationEvent WITHIN 60 WHERE TargetInstance ISA "Win32_Processor" AND TargetInstance.LoadPercentage > 99" could not be reactivated in namespace "//./root/CIMV2" because of error 0x80041003. Events cannot be delivered through this filter until the problem is corrected.

Then about 1 min later:

Publishing the Key Management Service (KMS) to DNS in the '' domain failed.
Info:
0x80070057

If anyone has any suggestions or need any more info, please let me know as I'm really trying to get this thing fixed so I can get back to farming in Diablo Thanks in advance for the help!

[Mr.Scott]

Set everything back to stock and try it. Eliminate the OC first. Then you'll know if it's a hardware or software problem.
To me, it looks like unstable OC or high temp. There are only a couple things that will lock Memtest or cause spontaneous re-boots. Bad OC, heat, or power.

[xipetotec]

That would initially be my thought, too, but I've been running the OC for 8 months without issue and seems odd that it would cause random problems now. In any case, I was able to get a BSOD, here's the pic below (sorry it's a little fuzzy, took with phone). Does this help with anything?

[Mr.Scott]

Generic BSOD. Could be anything from a corrupted OS file to a bad drive.
Set it back to stock and see if it BSOD's.
Sometimes OC's aren't as stable after time. Degradation you know.

[xipetotec]

Kk, I'll give it a shot and see how it goes. I'll post back when I know more. In the meantime, if you can think of anything else I can try or any other tests I can run, just let me know and I'll give it a shot.

[PommieB]

I agree with Mr Scott, you need to get rid of the overclock and do some testing, it could be the hardware will no longer handle the OC it did when it was new.

You need to post some temps including the rooms ambient temperature, telling us the temps were manageable doesn't mean anything, I float around a lot of forums and there idea of a manageable temp is 80C plus, they believe that that's a safe temp because Intel tell's them there cpu is good to 110C, what Intel fails to tell them is most motherboards won't last too at those temps.

[DarkFox]

slightly unstable OCs can cause data coruption, but its like russian roulette. You never know what file or when. Your 8 months could be the "luck of the draw."

After its back to stock, if your still BSODing, consider wiping your hard drive. If its data coruption, taking the OC out will not fix it, just prevent more.

[xipetotec]

Yup, still had a BSOD last night, so OC wasn't the sole cause at least. I'm guessing that wiping my drive and reinstalling is my next step? I was really hoping to avoid that, but I suppose it has to be done.

As for temps, when it was OC, I was seeing load temps about 57-58 C when running Prime 95. I only ran it for about 30 min, but I figured that if something was causing it to overheat, that running Prime would push it at least that fast, but I suppose it could have been data corruption.

[DarkFox]

Originally Posted by xipetotec

Yup, still had a BSOD last night, so OC wasn't the sole cause at least. I'm guessing that wiping my drive and reinstalling is my next step? I was really hoping to avoid that, but I suppose it has to be done.

As for temps, when it was OC, I was seeing load temps about 57-58 C when running Prime 95. I only ran it for about 30 min, but I figured that if something was causing it to overheat, that running Prime would push it at least that fast, but I suppose it could have been data corruption.

My guess would be yes to the first part. Just to confirm the second part though, did you re-test lately or is that when you first made your OC. Overtime dust and TIM aging as a big effect.

[xipetotec]

Originally Posted by DarkFox

did you re-test lately or is that when you first made your OC. Overtime dust and TIM aging as a big effect.

I retested a few days ago when I started having problems. Those temps are a bit higher then when I first did the OC 8 months ago, but still shouldn't be too hot.

I just reinstalled Windows last night and got really worried when I had a BSOD shortly after install while trying to run some updates. When it came back up, I manually updated my SSD firmware to 000F and it ran all last night without BSOD (which it hasn't done all week) so hopefully I found the problem. I'm still in the process of getting everything set back up, but looks like it was just my SSD firmware that was causing the problems, although not sure why that would freeze memtest, but oh well. I'll post back tonight once everything is reinstalled and I'm able to game again and see if it freezes since that was the biggest problem.

[Slizzo]

SSDs can cause issues like this if they're failing. I have a friend who has a failing SSD and he gets BSODs randomly quite often.

[DarkFox]

Suprising for a Cruicial, everyone raves about there stability. There are bound to be failures in all things though. Hope thats it for you and good luck.

You didnt by any chance default your bios completely did you? By that I mean when you installed windows the controller should still have been in RAID or ACHI mode.

[xipetotec]

Well, it appears that fixes the problem, everything has been running great for over 24 hours, I'll post back if I have any more issues. Gonna start over with the OC and see how that goes, maybe not quite as aggressively as before.

And yea, I did default bios, but luckily, I remembered to go in and change it to ACHI, lol. So shouldn't be any problem there.

[PommieB]

SSD's are blamed for a lot of problems that have nothing to do with them, the problem more likely a user problem than the ssd, that's why the administrators go to so much trouble having people try different things prior to rmaing the drive, it's usually not the drive, if it is it should be replaced.