System crashes and BSOD

[adamsonm]

I've been having regular crashes for about a month now. Sometimes the computer just freezes completely, with me having to reset the whole PC by the plug. Other times I get a Blue Screen Of Death with the following message

'a clock interrupt was not received on a secondary processor within the allocated time interval....'

I have been running my system with an over-clock for just over a year now. This is the first time I have experienced some sort of a crash. I ran stress tests back then to see whether the over-clock would be stable, and it passed all tests. The BSOD suggests, at least from looking on the internet, that its a CPU problem. So I ran another stress test yesterday for about 5 hours. No crashes or BSOD.

I put the CPU back to the factory default clock and it actually got worse. The computer would just crash at windows login screen or just freeze completely again. So I am currently running a very low over-clock to be able to use my PC.

One thing I noticed is that the PC crashes more often when I'm about to watch a video on YouTube. After a few restarts (via the plug) the PC would work normally, I can watch videos on YouTube, play games with no crashes.

It definitely seems random to me, however if it was the CPU wouldn't have the stress test failed??

Maybe it's the Motherboard??

Any ideas?

[kalniel]

Suggests voltages from the PSU to me, rather than CPU or motherboard. Try upping voltages and seeing if that stabilises things.

[najiro]

To help you determine the problem, please answer the following questions:

1) How old is your system?
2) Specs?
3) Recent changes done before the problems occur.

[adamsonm]

@kalniel - Do you mean upping the voltages of the overclock??
@najiro - I built my system in the summer of 2010. Specs are as follows

Motherboard: GBYTE GA-X58A UD3R
CPU: Intel Core i7 920
Memory: Corsair Memory XMS3 8GB
Storage: Sandisk 128 Gb SSD AND 2TB Samsung HD203WI Spinpoint F3EG AND 1TB Samsung HD103SJ Spinpoint F3 AND 2TB Samsung HD203WI Spinpoint F3EG
Graphics card(s): ATi XFX 6870
PSU: Corsair TX650W
Case: Lancool PC-K62 Mid tower
Operating System: Windows 7 64 bit
Monitor(s): Viewsonic VX2260WM

I started running a dual monitor on my system this summer, in the middle of June to be precise. The crashes started occuring from 11th July onwards. So the dual monitor is the only change I made in this past year.

[kalniel]

Voltages of stability related components - could be vCore, could be VTT, could be dimms etc. Rationale being if a mild overclock actually helps stability it could be because by overclocking you're increasing any auto controlled voltages.

[adamsonm]

These are my temps and voltages at idle. The max temperatures might be a bit high but thats because I recently did a stress test to test hardware.

[stilkun]

Have you had a look through the minidumps? You can use BlueScreenView to examine the dumps which tells you the culprit is the cause.

[adamsonm]

@stilkun I have downloaded and used BlueScreenView. Unfortunately no dumps have been created for the BSOD 'a clock interrupt was not received on a secondary processor within the allocated time interval....', although I have seen other dumps for other errors. I will look today to find out more about the errors. Like I said though, the program doesn't show the dumps for the BSOD about the clock interrupt.

[Phage]

Try WhoCrashed. It's another dump analysis tool.
http://www.resplendence.com/whocrashed

[najiro]

ohh.. a 1366 platform.. i have no experience with those and it they behave like 1155 builds. Hmmm.. why would a dual monitory setup cause BSODs.. I don't think that's the culprit.. something must be ageing in the system.

[mikeo01]

That type of error usually indicates some sort of CPU error or voltages.

I haven't had experience with overclocking Intel chips so not sure whats safe for them, but was your overclock actually stable to begin with? Have you exceeded temperatures or safe voltages for this chip perhaps? The motherboard seems capable but the error sounds like the chip isn't getting enough power, whether a cores gone dodgy (doubtful), VRMs not stable or power supply not outputting enough power.

As mentioned above make sure the PSU isn't the problem, hopefully you haven't overloaded it at some point causing a slight degradation of the PSU to output power.

It seems current/power related, or in an unlikely event a damaged CPU due to high/low voltages and temperatures.

[adamsonm]

Thanks for the replies guys, I really appreciate it. Unfortunately I am still having these random hangs/locks. Again, it happens more often when I have just started watching a video on YouTube. I managed to put the CPU to Fail Safe settings, however that didn't help. I also increased the voltages for a few settings, within the motherboard start up screen, however that didn't help either. Voltages seem fine to me. I checked them through the bios screen. I don't trust programs to show me the voltages because they are never accurate.

Anyway, I will run another stress test on the CPU using a different stress testing tool. I will also have a look at the other dump analysis tool.

I'll keep this updated.

[kalniel]

Command rate for RAM? Make sure 2T.

[NecronomicoN]

If you're prepared to try something odd, uninstall Flash Player. That should force YouTube to use HTML5 (for most videos). If the crashes subside, it indicates a bad video driver. If that's the case, rollback.

I only mention this because it happened to me a few weeks ago and drove me nuts until I traced it to the GPU driver via Flash content.

[kalniel]

No need to uninstall flash to do that, you can just opt into the HTML5 trial here:
http://www.youtube.com/html5?gl=GB

[adamsonm]

From the mini dump analysis tools, I can see that the errors that do happen all point or I mean the majority of the crashes point to ntoskrnl.exe.