Bsod

[stevieray]

I just had a curious experience.

Upon start up, everything proceeded normally until the sign in screen... which disappeared after a split-second, and a BSOD popped up instead. It said something about a "Paging file in a non-paging area".

It gave me the option of starting Windows normally, or running a repair. I chose the repair option. That led to a second blue screen, telling me to uninstall any new devices. I didn't, because there were no new devices. I had to do a hard shut down to restart it.

After that, it brought up the usual black screen, saying "Windows didn't shut down normally...". From there, it seems to have started up correctly. Everything is where I left it earlier today.

Question: Is there any way I can find out what the heck happened? Some sort of error log I can check and decode?

Thanks

 

[torrentg]

Sure, there's logs for your system.

Type "event" without quotes in the start menu search box. Open event viewer that shows. Now have a look at the windows logs.

 

[stevieray]

Thanks Torrentg.

I tracked down the bug check from the relevant event (I think) in the minidump. I want to take a look at it, but I can't get Notepad to open it... it gives me an "access denied" message. I took ownership of the minidump, opened notepad as an administrator, and I still get the "access denied" message.

What should I open this file with?

Edit: Also tried opening it with WordPad. Still getting "access denied". Changed the permissions for the entire minidump file to "Everyone", but still cannot access it. Grrrr....

 

[stevieray]

BSOD Again

Well, I just got the BSOD again. I was simply browsing at the time.

I wonder if this is related to the Gadget crashes I've been experiencing today? Every hour of so, the gadgets shut down, and my pc looks online for a fix. They restart a few seconds later.

I had no crashes at all until today... now I've had two in a few hours. The only changes made to my system in the last day is the Windows 7 updates sent out last night. They offered Silverlight and Windows Live Essentials with the updates.

Hmmm. Think I'll uninstall Live Essentials... see if that helps.

 

[stevieray]

Crash #3. Went to the bathroom, came back to find the "Windows didn't close normally" screen.

Live Essentials is uninstalled... guess it wasn't that.

 

[stevieray]

I'm up to five crashes now... all since 5:00 a.m. this morning.

During the last crash I noticed the reason given for it -- "Bad Pool Caller". Not very helpful... could be anything from a bad device to bad ram.

I still cannot access the minidumps. I have granted myself full access, for both the minidump folder and each file in it, but I still get "access denied" messages. I checked the "properties > security" tab, and I have the same checkmarks as System.

I downloaded WhoCrashed and ran it. It cannot install the debugging tool... it crashes and gives me an error message (2203).

I tried manually downloading the debugging tool and symbols. The Itanium version won't install (as expected... I'm not on an Itanium machine), and the regular x64 can't access the minidumps either... "access denied" again.

I can copy the minidump files to my desktop, but the debugger can't read them there... it gives me a "xxx isn't a windows file name" message. I can't copy them to Notepad... the "paste" function stays grayed out.

I ran sfc /scannow. It repaired some things, but couldn't fix them all... and gave me the example of CBS.log. Strangely, I could bump myself up to Admin status for those files, and I checked them out in Notepad. Judging from what I could make out, that was the corrupted debugging tool that WhoCrashed tried to install.

And to further confuse me, I checked out CPU-Z. I don't know much about it, having learned of it only two days ago, but it looks to me like my CPU spends most of its time running at 2/3 speed (1600Mhz instead of 2.4 Ghz), and my RAM is running as PC2-6400, instead of 8500... the memory slot voltage is back in the mb's default setting of 1.8 volts instead of the 2.0 my RAM calls for.

I'm at a loss here... don't know what to try next. I could use some ideas...

 

[torrentg]

This should read about the crashes:

http://www.microsoft.com/whdc/devtools/debugging/default.mspx

 

[Hammerhead786]

Check your bios and see what the CPU is reported as in there. My CPU was getting very hot and it was running like a dog with two legs. Bios temps were quite high. After some fresh thermal paste on the processor cooler, the processor was back to running at normal speeds and temperatures dropped back to half. Let us know how you get on.

 

[usasma]

I find that most crashes are resolved after looking at what you've done prior to when the crashes started. It's usually something that you'll remember that seemed a bit off when you did it (such as installing or removing a program/update).

If you have the correct Debugging Tools installed for your system, you can run the debugger and then drag a dump file from the desktop onto the debugger window (and it should launch. More instructions on how to use it here: Diagnosing Bsod's (originally posted 02Oct08)

Alternatively, you can select the minidumps on your desktop by pressing "Ctrl" and then left clicking on each one. Then right click on them and select "Sent to...", then select "Compressed (zipped) folder". Then upload that zipped file along with your next post and we'll have a look at them.

While waiting, I'd suggest that you check your RAM and hard drive. Here's some links:
RAM: Bootable Memory Test Diagnostics
Hard Drive: Bootable Hard Drive Diagnostics

 

[stevieray]

usasma: Is that your site? Nice! Its going in my bookmarks. I had no crashes at all until after the updates for rc1 on Sunday night. You've given me enough tests to keep me busy for a while... thanks!

hammerhead: I haven't spent much time at all looking at the bios for this pc. I wouldn't know what was normal and what was wonky. I have one of those dual bios Gigabyte mbs designed for overclockers... it has a billion user-changeable settings... probably all set on [auto] right now. I was planning on figuring it all out in the future... I guess the future arrives today!

torrentg: Thanks for the link. I'll use it to double check the debugging tool I installed... make sure I've got the right version. But until I figure out why I'm denied access to the minidumps, I can't use the tool. I guess I can temporarily disable UAC... hmmm...

Good thing I have this week off from work! Gonna be busy! Heh

 

[Anthony]

Have you tried removing some of that ram?
Try it with one stick of ram see how you go, then start adding the rest one at a time, Good Luck

 

[Hammerhead786]

stevieray, check your cpu speed and temperature in the bios as shown here:

 

[stevieray]

I had a few obligations in the meat world to attend to... I'm back home now, running all of the tests y'all suggested. I'll post the results in an hour or two.

Heh. It looks like a computer store in here. My desktop's got its memory pulled out; my laptop (vista) is installing the new service pack and rebooting like a madman... I had to set up my old xp pro desktop on my bar!

 

[stevieray]

stevieray, check your cpu speed and temperature in the bios as shown here:

How do I use the snipping tool in BIOS... before I let Windows open?

 

[stevieray]

More Info

I finished running the memtests on my RAM... no errors found on any card or in any slot.

I tried to run the recommended HD test, but I couldn't find a WD test that'll run in Win7.

I disabled UAC so I could access the minidumps. Ran WhoCrashed, and here are the reports:

View attachment 12119

View attachment 12120

I googled the problems... narrowed it down to RAM... or a video driver... or a processing error... or bad karma...

I checked out CoreTemp64. The temps are a bit too high, but not close to the limit. They should be 10 degrees lower under these minimal loads. I keep the core temps on display in my task bar, and I've never seen it hit 50... its usually in the 30's, with spikes up into the mid 40's under medium use.

View attachment 12121

I'm gonna poke around some more... see what else I can find.

 

[Hammerhead786]

Your processor speed is definitely not right. Here's mine using CPU-Z:

 

[stevieray]

Your processor speed is definitely not right. Here's mine using CPU-Z:

Yes. I noticed that late last night. It runs at 2/3rd speed about 90% of the time, and jumps up to full speed every 10 to 15 seconds. Notice the multiplier changes too.

View attachment 12129

View attachment 12130

Could this be some sort of power-saving "green" setting in BIOS?

 

[Hammerhead786]

Most certainly. If you look in the advanced bios features you should find the following:

 

[Elooder]

Funny...

...how you find your 35 degree temperatures to be "a bit high" I have a q6600 at work that, on stock speed, hovers around 40 and as it is now (oc'd to 3200) never goes below 45. Same goes for another of my work computers which is a q9450 (50 - 55 @ 3.2 Ghz) and my home i7 920 running at 3.5 Ghz... I think 35 C for q6600 on stock is pretty good... at least I've never seen it run cooler (ok, so I may be using crap coolers, but the one on the q6600 is the V-shaped scythe, the q9450 has a Zalman 9700 on, and the i7 has an ASUS "something 81" on it... can't seem to remember the name).

Are you using the drivers for your video card that Windows itself installs from Windows update? I've found that the current WHQL certified drivers that are available on nVidia's site at the moment cause my computer to crash EVERY time I log off a user... That's on a GFX 280)

 

[stevieray]

Elooder: I'm basing my view that the temps are a bit too high on, well, very little at all.

When I was weighing different CPU's before I built this thing, I went by the Intel-run user forums to see what people were complaining about. That's a habit I picked up while searching for av/as/fw suites -- magazine/website recommendations are good to know, but the real poop is on the user forums... see what they are talking about, and how responsive the staff is, and make your choice based on that.

Anyway, I noticed the intel gurus were generally running in the mid-twenties at idle, mid-forties at full load.

Looking around, it looks like most folks get core temps in the same range as us... higher than the Intel guys, but still within parameters. But still, I'm planning on replacing the stock cooler before I try any overclocking.

As to the video driver, the original one that came with the card is the most current, so I'm using it. nviddmkm8.15.11.8585. The device manager seems to like it... who am I to argue?