About 2 months ago I bought a new system and built it at home:

  • Mother board: XFX X58i
  • Processor: Core i7 920, using the stock cooler
  • Memory: 3x2GB Corsair DDR3 1600
  • Video card: NVIDIA GTS 250 (1GB)
  • Hard disk: 2x WD 500GB, 7200rpm

I have 2 screens plugged into the video card, and the system is connected to a 550W PSU. Nothing is overclocked.

After building the system, I stressed it a lot with Prime95 and rthdribl to check its stability. All my tests were perfect. So I reinstalled Win 7 x64 Professional and started using it normally.

The first week (2010-03-15) I got the infamous irql_less_than_or_equal BSOD. Ten days after (2010-03-24) I got another one. Then on 2010-04-09, 2010-05-04.

Since 2 days ago it became worse: I got one bluescreen per day! (2010-05-12, 2010-05-13, 2010-05-14).

I installed BlueScreenView to try to obtain some information, but I'm not able to extract any useful information apart from the bug check string (irql_less_than_or_equal), and that it was caused by ntoskrnl.exe (the first three at ntoskrnl.exe+71f00, the last 4 at ntoskrnl.exe+70600 -- which I suspect could be the same thing, as Microsoft could have patched this file in the mean time, so the address of the function causing it changed).

Then I stressed my memory sticks with memtest, they worked perfectly. After booting, I've stressed my GPU with FurMark and RTHDRIBL, everything was fine. Then I stressed the CPU with 4 instances of Prime95 while monitoring the temperature -- that never exceeded 85oC with the case closed --, everything fine. Finally I've stressed the whole system with HeavyLoad for a looooong time, everything worked just fine.

So, I have stressed most of the components of the system, but couldn't get any useful information from it.

Do you have any hint on what else can I do to find the culprit?

EDIT: I have a MyBook external hard-disk I forgot to mention. I've unplugged it for some days, and have had no bluescreens. The other day, I plugged it again, and I got a bluescreen. The external hd might be related to the bluescreens. Is there anyway to check if this is or isn't the case?

  • 85 deg C seems very hot for a CPU to me...but maybe the newer i7 CPUs can run hotter? I think the maximum temperature for my older Q6600 quad core intel is 80 C...
    – davr
    May 15, 2010 at 4:02
  • Yes, but I get this only under heavy load, and I almost never run on heavy load. Also, the BSODs are not related to the temperature I think because the system was not under heavy load when I got them.
    – Bruno Reis
    May 15, 2010 at 4:08
  • The temperature under my normal load is around 45, 50 C. The external temperature is around 27 C.
    – Bruno Reis
    May 15, 2010 at 4:10
  • I was running my Q6600 on VERY HIGH temperature without a problem. (Don't want to say stupid things but the fan stuck one time and it was running like that for a week. Even games were running without a FPS drop. (No, I wasn't at home. :)))
    – Apache
    May 21, 2010 at 5:56
  • @BrunoReis: You can mark your question as solved by answering it yourself and marking the solution as accepted. It gives you some extra reputation... :) Apr 9, 2011 at 14:57

1 Answer 1


Problem solved

These bluescreens were caused by some bad code written by Shrew Soft, for their VPN Client. I discovered and solved this a long, long time ago -- and since I've never, ever again, installed software from them -- so I don't remember the exact version. I must say that at the time I googled a lot and found people complaining about blue screens related to the Shrew Soft VPN Client, but the reports were about some older version. It looks like the problem, if fixed now, has taken a very long time to be fixed.

To find the cause, I used a debugger from Microsoft, and I found a call stack going to some DLLs from Shrew Soft.

After removing the software, I have never again seen a blue screen. I'm almost starting to miss them.

EDIT: related question

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .