Random Windows 7 Lock up

[willielikesmonkeys]

I have a custom built Win 7 machine that is using older hard drives in raid 0 and an unnamed psu ( http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=1273337 ). The machine randomly locks up and become unresponsive (no bsod) and requires a force reboot. The same problem was to be had when it was on windows 8, so I'm pretty darn sure its hardware failure. So does anybody think its a problem with a bad PSU or the HDDs? 

[A/C]

Based on my previous experiences I'd look for RAM, RAID, and CPU problems before I look into PSU problems. Troubleshoot RAM and your raid disk before anything else. Smells like a memory leak to me, but I could be wrong, anything could be causing the lockups, even the mobo, or gpu...

[willielikesmonkeys]

Based on my previous experiences I'd look for RAM, RAID, and CPU problems before I look into PSU problems. Troubleshoot RAM and your raid disk before anything else. Smells like a memory leak to me, but I could be wrong, anything could be causing the lockups, even the mobo, or gpu...

I recently added a new GPU (Radeon 7850) and the problem wasn't effected at all. Would downclocking the RAM help with a memory leak? Its running at the stock rating ( 1600 MHz ). And I've had the machine running in RAID and on both of the individual drives, it didn't seem to make a difference either way I still had the problem. Also something I forgot to mention is that whenever I'm running Linux (I've used Arch and Ubuntu on this machine) I never have this problem.

[A/C]

 

I recently added a new GPU (Radeon 7850) and the problem wasn't effected at all. Would downclocking the RAM help with a memory leak? Its running at the stock rating ( 1600 MHz ). And I've had the machine running in RAID and on both of the individual drives, it didn't seem to make a difference either way I still had the problem. Also something I forgot to mention is that whenever I'm running Linux (I've used Arch and Ubuntu on this machine) I never have this problem.

 

hmm... if the ram speed was the problem, windows would most likely crash, not lock up, as would linux, so I don't think downclocking the ram would help there. If you have multiple ram sticks in your mobo try running windows with only one and see if it locks up. If it does, replace the removed stick and take out the one that you didn't remove the first time, just so you eliminate all of the ram sticks.

Do you run linux off the same raid drive?

[willielikesmonkeys]

Yes I do run Linux off the the same raid array. No problems with that. And It's 2 sticks in dual channel I'll try running single and see what happens. But that's the really weird part, it doesn't matter if it's under load or not. It happens if I leave the machine on over night seeding torrents or if I'm playing Crysis, doesn't seem to matter.

[willielikesmonkeys]

Quick update, I opted to first try turning down the clock to 1333 MHz and left the machine overnight seeding, where normally this would have meant the machine crashed it did not. Now to try Crysis and benchmark this thing. Will keep updating as things happen or (hopefully) don't.

[willielikesmonkeys]

So after a 6 hour benchmark of Crysis, there were NO crashes. Hurray! Still not ready to call it case closed though but if I don't post to this thread again, assume the problem is solved.

[willielikesmonkeys]

Alas I spoke too soon, If nothing else it did seem to prolong it so I opted to turn it down again to 1066. When it did crash though I had left the machine idle when I left for 30 mins and came back to it unresponsive. 

[A/C]

Alas I spoke too soon, If nothing else it did seem to prolong it so I opted to turn it down again to 1066. When it did crash though I had left the machine idle when I left for 30 mins and came back to it unresponsive. 

 

any oc on the machine?

 

tried running just one ram stick?

[willielikesmonkeys]

any oc on the machine?

 

tried running just one ram stick?

No everything it stock clock speeds, even turned off turbo. And I'm running one ram stick right now after the 1066 failed.

[A/C]

No everything it stock clock speeds, even turned off turbo. And I'm running one ram stick right now after the 1066 failed.

 

k, report back

[willielikesmonkeys]

So it's been running for a day in single and so far no crashes with about 20 hours uptime. So far so good but I may try running it in single with the other stick just for the hell of it after I've had this thing with over 50 hours uptime. Also, just curious, but ram is fine at the same temp's as a CPU right? Because reaching in and feeling its MAYBE 45-50 C leaning towards 45 (about 110 F) just by feeling the heatsink on the RAM.

 

In HWMonitor TMPIN0 is at 38C, TMPIN1 is at 43C, and TMPIN2 is at 40C. Though I'm not sure which sensor is the CPU socket/Chipset/RAM socket. MSI ClickBIOS reports CPU is the 40C. 

[willielikesmonkeys]

I've been running single stick for 2 days now with an uptime of 2:08:34:00 So I think can safely say this first stick is okay, at least within reason, it was running at 1333 even though its rated at 1600. So I'm going to try switching it with the other stick just to be sure, you never know if its something weird with the motherboard since it was pretty cheap. 

[willielikesmonkeys]

So THIS is weird. I tried the second stick and now I'm at 1:23:29:00 uptime. So the RAM is good? So it might just be an issue with the motherboard after all... or CPU... Well this complicates things. Especially since I don't have any known good parts to test this with.

 

POST: just for shits and giggles I put both sticks back in and adjusted the cpu cooler a bit. I also left the side panel off. We'll see what happens.