Solved BSOD each time my computer starts..

December 17, 2012 at 21:22:21
Specs: Windows 7
I get a BSOD each time my computer starts.. but then the second time around it starts fine. It's really strange and annoying. I'm attaching a rar file of all my dump files. I've checked myself and I noticed ntoskrnl.exe coming up a lot.

I don't know enough about checking for dump files.. and whoever figures out my problem could you care to explain how you did it, so I may learn.. Thanks!

http://www.sendspace.com/file/4enole


See More: BSOD each time my computer starts..

Report •


✔ Best Answer
December 20, 2012 at 08:51:42
mmm As I commented/suggested earlier.... Memory can sometimes produce this problem... And then also a drive starting to fail. Of the two I'd be considering RAM first... (Although avg may also be a player here...) considering RAM issues first...

In a somewhat mechanical sense this is how I suggest it manages to boot the second time...

First attempt it fails due to dodgy/faulty segment/s in RAM; critical files ntoskernel.exe and other early load files being not infrequent casualties.... Rebooting (warm/soft start - not powering down fully in the process - see more on this below) some of the critical start-up files may/now go into another area of RAM (that is OK) and thus allow you to boot OK...

Changing RAM in the past has resolved this for me on a couple of occasions (but to be fair - not always).

If you switch off a typical atx motherboard style desktop system (anything from late mid-late 90s onwards?) - but leave mains volts going into the system (i.e. mains outlet is still switched "on"), then there are active voltages running around the motherboard components - including RAM... With volts still present RAM holds its charge (i.e. information may be - or is - retained, and thus those areas are not available for other use - as in the M$ memory leak issue; which means information now has to go elsewhere (if it can). Removing power completely - no volts going into system at all, then no volts running around and thus RAM is also discharged... This resolves the M$ problem of RAM being slowly "used up" as it were by the Office applications... (aka memory leak if I recall properly?) which causes M$ apps to run slower and slower until more or less grinding to a halt... Killing the system completely removes (clears/discharges) RAM modules fully and thus max available for use... In the case of a laptop, even if you remove the mains/psu adapter there is still a battery involved... Remove the battery too for about thirty seconds or so to again discharge/clear RAM... This was a standard "kwik fix" for all our workstations and laptops that were grinding to a halt running many consecutive sessions of M$-Office... - without a full power down/cold-boot between users, which was often not practical... (This was in a 24x7 broadcast news environment)

Quite by chance the above led to this observation...If you cold boot a system and it fails to boot fully, but will on the second try - usually via a (soft boot) restart - sometimes referred to as a warm boot... - what seems to happen is that boot/start-up information is written (on second try) to sections of RAM that are OK (i.e. not to the same area/s of RAM as on first try); as the earlier areas are now already flagged as in use.... If you cold boot for the second try - as in remove all power and the battery too if applicable (a laptop) - then frequently you will not reboot OK on the second try...

I appreciate the above explanation etc. is not exactly hard science; more anecdotal based on observation and experience... However it has stood the test on more than a few times... Replacing the RAM modules later usually resolved the boot issue thereafter...

I have mixed feeling re' memory test utils; have found the best test being to replace suspect RAM and see what happens... This approach is usually adopted to in large IT departments; as RAM tests take time...?

I also have had serious issues with avg in the past; it crippled an otherwise fine ME installation...; and frequently stalled one 98SE system. Removing avg and all was well... I replaced it with avast... which was fine (then...). So if you can boot, be it into safe or normal mode..., suggest you remove avg; then see how things go? When you set about removing avg... first disable system restore; and re-enable it when all is well.

http://www.howtogeek.com/howto/3187...

Also you say this has been going on for about two months... Do you have any restore points previous the start date of the problem; and if so have you considered going back to one of those?

Even though I am orbiting around RAM a lot here... I'd be much inclined to remove avg first and see how it all fares... If the problem then goes away... replace avg with either M$ Security Essentials package (which I find fine); or avast which XpUser and others currently favour...; as either ought be better than avg...?



#1
December 18, 2012 at 03:50:05
I would start by updating all device drivers.

Report •

#2
December 19, 2012 at 05:32:12
give us the EXACT error number of the BSOD so we can help you. Your link is not good.

Some HELP in posting on Computing.net plus free progs and instructions 7 Golds


Report •

#3
December 19, 2012 at 08:46:39
mmm... On occasion I have found a duff (faulty - or starting to fail) memory module can produce this effect...

Not always the cause of course; but no harm to try...

Presuming two or more RAM sticks, remove one (leave at least 128Meg - preferably more... as a single stick in place) and see what happens when you boot... Swaps sticks in/out so as to test any/all sticks...?

Is this a laptop or desktop system (guessing a laptop), and what make/model? Regardless... if/when you remove RAM sticks, clean their contacts with a soft pencil eraser (remove any debris that may occur as you do so from contact surfaces; and when re-installing them, ensure they are securely inserted etc...

One other approach... If you have full windows disk (not a recovery disk) that allows you to boot up into usual setup mode etc... does that allow you to get to the usual setup etc. menu and so? If so... a repair install might allow a fix...

However - as data etc. is usually the first issue for many when system fails to boot OK... It might be worth to save all that "stuff" to external media asap; then try to fix what ails the system. Use a Linux disk - usual variants are around and downloadable as freebie ISO - which burn the disk and boot with that... Then you can access the data and thus save it all elsewhere. Debian, Knoppix, Ubuntu are the usual three that come to mind? And of course if/when you do manage to boot on second go - you can save all that "stuff" then to external media...

And also verify the copies are readable on at least one other system; just in-case any one disk/copy isn't all it orta be...?

I have known this issue to arrive with an HD that is starting to fail - but not yet given up the ghost... Which is why I suggest data recovery first - if possible... There are HD test utils various out there to check the HD integrity anon... Where I used to work a while back we had a spate of IBM drives that gave this problem... It was just one batch of kit that had the same model/spec drive etc...; reloads etc. of OS/apps. didn't remove/prevent the issue re-occurring at a later date; but when those drives were replaced - no more problems...

Decoding/decyphering BSOD data is a wee bit of an art; unless you're really across it... Sadly I'm not these days, but as XpUser suggests, post details more fully here and possibly he (and/or others) can decode according...?
Also... can you boot (first time) if/when select to go Safe-mode?

Also ensure you don't have anything external attached to the system when booting; though I doubt this is is the cause/problem here; As "Paul... - post 2" suggests - drivers various "might " be an issue; but again I have seldom found that to be the case (at least with anything from '95 - XP - not that experienced - yet - with win-7).


Report •

Related Solutions

#4
December 20, 2012 at 06:17:27
I've updated all my drivers, and I still have the problem. This is a consistent problem, I can ALWAYS boot up the second time. This problem has been going on for at least 2 months. I can always enter safe mode, but I've never tried doing it on my first boot up before it restarts. I'm using a desktop PC, and I've tested my memory with a diagnostic tool, which is fine.

When I take a look at my dump files using BlueScreenView there's a lot of problems focused around ntoskrnl.exe and avgldx64.sys... I'm guessing avgldx64 has to do something with my AVG antivirus.. the other file I have no clue. I'm not quite sure what to exactly do here, I could uninstall AVG.. but for the other file I don't know.

Here is a re-upload and a newer version of my dump files. http://www.sendspace.com/file/gmlc5a


Report •

#5
December 20, 2012 at 08:03:02
What happens between the two boot attempts? Does windows offer the boot options screen, or go into mandatory chkdsk?

Report •

#6
December 20, 2012 at 08:51:42
✔ Best Answer
mmm As I commented/suggested earlier.... Memory can sometimes produce this problem... And then also a drive starting to fail. Of the two I'd be considering RAM first... (Although avg may also be a player here...) considering RAM issues first...

In a somewhat mechanical sense this is how I suggest it manages to boot the second time...

First attempt it fails due to dodgy/faulty segment/s in RAM; critical files ntoskernel.exe and other early load files being not infrequent casualties.... Rebooting (warm/soft start - not powering down fully in the process - see more on this below) some of the critical start-up files may/now go into another area of RAM (that is OK) and thus allow you to boot OK...

Changing RAM in the past has resolved this for me on a couple of occasions (but to be fair - not always).

If you switch off a typical atx motherboard style desktop system (anything from late mid-late 90s onwards?) - but leave mains volts going into the system (i.e. mains outlet is still switched "on"), then there are active voltages running around the motherboard components - including RAM... With volts still present RAM holds its charge (i.e. information may be - or is - retained, and thus those areas are not available for other use - as in the M$ memory leak issue; which means information now has to go elsewhere (if it can). Removing power completely - no volts going into system at all, then no volts running around and thus RAM is also discharged... This resolves the M$ problem of RAM being slowly "used up" as it were by the Office applications... (aka memory leak if I recall properly?) which causes M$ apps to run slower and slower until more or less grinding to a halt... Killing the system completely removes (clears/discharges) RAM modules fully and thus max available for use... In the case of a laptop, even if you remove the mains/psu adapter there is still a battery involved... Remove the battery too for about thirty seconds or so to again discharge/clear RAM... This was a standard "kwik fix" for all our workstations and laptops that were grinding to a halt running many consecutive sessions of M$-Office... - without a full power down/cold-boot between users, which was often not practical... (This was in a 24x7 broadcast news environment)

Quite by chance the above led to this observation...If you cold boot a system and it fails to boot fully, but will on the second try - usually via a (soft boot) restart - sometimes referred to as a warm boot... - what seems to happen is that boot/start-up information is written (on second try) to sections of RAM that are OK (i.e. not to the same area/s of RAM as on first try); as the earlier areas are now already flagged as in use.... If you cold boot for the second try - as in remove all power and the battery too if applicable (a laptop) - then frequently you will not reboot OK on the second try...

I appreciate the above explanation etc. is not exactly hard science; more anecdotal based on observation and experience... However it has stood the test on more than a few times... Replacing the RAM modules later usually resolved the boot issue thereafter...

I have mixed feeling re' memory test utils; have found the best test being to replace suspect RAM and see what happens... This approach is usually adopted to in large IT departments; as RAM tests take time...?

I also have had serious issues with avg in the past; it crippled an otherwise fine ME installation...; and frequently stalled one 98SE system. Removing avg and all was well... I replaced it with avast... which was fine (then...). So if you can boot, be it into safe or normal mode..., suggest you remove avg; then see how things go? When you set about removing avg... first disable system restore; and re-enable it when all is well.

http://www.howtogeek.com/howto/3187...

Also you say this has been going on for about two months... Do you have any restore points previous the start date of the problem; and if so have you considered going back to one of those?

Even though I am orbiting around RAM a lot here... I'd be much inclined to remove avg first and see how it all fares... If the problem then goes away... replace avg with either M$ Security Essentials package (which I find fine); or avast which XpUser and others currently favour...; as either ought be better than avg...?


Report •

#7
December 20, 2012 at 09:58:27
The dumps show a lot of crashes caused by trying to read unreadable memory, or illegal instructions. Seeing as this happens when you first start up, I'd suspect the memory has a flaw that only surfaces when it's cold.

If you're quick about it, a memory tester might be able to spot the error if you boot into it from a cold system. Otherwise, you might just want to swap out RAM to see if you can spot the bad stick.

Optionally, you could just ignore it until your system starts crashing after it has warmed up.

How To Ask Questions The Smart Way


Report •

#8
December 20, 2012 at 14:46:03
Again, not too scientific, but quick to try:

1. Clean the RAM edge connectors with a pencil eraser then pop the sticks in and out a few times to clear any oxide off the sockets themselves.

2. Already mentioned but just expanding the procedure a bit. Unplug the computer from the mains, remove the main battery too if it's a laptop. Hold the power button in for about 20 seconds, see if it helps when put back together.

Always pop back and let us know the outcome - thanks


Report •

#9
January 3, 2013 at 15:48:09
Yep, it was the RAM. I gave it a couple of days of testing before I responded back here. Funny thing is.. the moment I removed the RAM and tried putting it in a different spot, my computer wouldn't boot to windows at all.

Thanks everyone!


Report •

#10
January 3, 2013 at 15:54:38
Re #9
That symptom (fails when moving RAM) is often oxidized edge connectors (#8).

Whatever, glad to hear you've fixed it and thanks for popping back to let us know.


Report •


Ask Question