Various Random BSOD's after reapplying thermal paste

[Carter15]

Hello everyone,

so the thing is that Im getting random bsod's, most of them don't even show the blue screen and pc just "stops" and restarts, sometimes bios doesn't show up (have to turn off power), sometimes pc boots again.

It all started about 1 year ago when I needed to reapply thermal paste because of high temps on graphics card and processor, temps were fine after reapplying. I also reseted bios (switched jumper on mb and took off battery while manipulating with cpu and graphics). At first boot I've set up bios and booted to win. All ok.
But one problem showed up next day. After turning on pc bios detected "new" cpu and showed this message "New CPU installed! Please enter setup to configure your system." "Press F1 to enter setup" "Press F2 to load default values and continue". I just entered setup, checked config and saved it. Next day this showed up again, entered setup and ESCaped right away.
So I tried to search for some solutions, I've read about trying to reseat cpu heatsink (so I did) and turning off the error messages in bios (but that won't solve my problem, so I didn't do that) and some other tips that didn't seemed useful for my problem. After reseating it the problem with message dissappeared. After some time it came back. So I just reseated heatsink again.
This is happening over and over again for about a year now. Sometime it lasted fixed 1-2 months, sometime 1 week, 2 days, etc, etc.

Now to the issues that brought me here actually.
About half year ago or more, I started to get 0x00000124 STOP bsod's. As I figured out they are caused by that bios message with new cpu. After that year I became lazy to reseat heatsink all the time, so I was just skipping it by F1 in and ESC'ed out and when I was doing that, those 124 stop bsod's were happening. When reseated, message dissappered and no bsod's happened.

About a month ago or so, bsod's became more serious.
I started to get various bsod's/crashes where suddenly screen goes black and pc restarts (sometimes bios shows and win boots, sometimes it doesn't as I said before). Reseating heatsink doesn't help anymore in preventing bsod's. In addition to the 124 stop's I'm now getting some new different bsod's, that I never had before.
Still most of the times blue screen doesn't show up and also most of the times minidumps are not created. So I have few of them at the moment (will add when new are created).
Last week or two when Win boots up, pc pretty much everytime crashes in like 5-10 minutes after booting. After that other bsod's/crashes are completely random. They happen while playing games, while browsing internet or when pc is idle, so they are not caused by doing something.
I also have photos of about 3 bluescreens that stayed on screen till I hard reseted pc, so I can send them if needed.


For information, I've never had single bsod or crash before reapplying the thermal paste. PC will be 5 years old in 3 months. So about 4 years without a problem.

Anything in PC is NOT overclocked in any way, everything is on stock settings and controlled by bios/drivers.
BIOS is updated to latest version, graphics card drivers also.
In fact, everything that was needed to be installed 2 years ago, when I was reinstalling Windows, I downloaded the latest versions from internet (I mean the hardware drivers now, like chipset, soundcard drivers, etc.). There are no newer versions now as the hardware is old.

I tried memtest86 (2 tests with over 10 passes), cpu and graphics stress tests (no bsod's/crashes druing tests), CHKDSK, SFC (didn't repaired anything), everything completed without any errors or problems.

Im including zip from DM Log collector and in separated zip some dumps from May.

Thanks for help.

Im sorry for this super long text and my english too.

TLDR
Im having problems with a lot of random bsod's/crashes after reapplying thermal paste, during last month more of them than ever. For more info read text above

 

[Carter15]

Uploading one new dump.
Yesterday my pc crashed 7 times, none of the crashes made dump, all of them in event log are Kernel-Power, Event ID 41, Task 63.

Any help appreciated, thanks

 

[Arc]

event log are Kernel-Power, Event ID 41, Task 63.

That means abnormal shutdown. It may be a BSOD or may not be, too.

Checked your recent crash dumps, all different bugcheck codes.

[COLOR=Red]BugCheck 3B[/COLOR], {c0000005, fffff800033693e5, fffff88003d60000, 0}

Probably caused by : ntkrnlmp.exe ( nt!EtwpAccessCheck+7d )

Followup: MachineOwner
---------

[COLOR=Red]BugCheck 101[/COLOR], {31, 0, fffff88002fd5180, 3}

Probably caused by : Unknown_Image ( ANALYSIS_INCONCLUSIVE )

Followup: MachineOwner
---------

[COLOR=Red]BugCheck 124[/COLOR], {0, fffffa8003ad08f8, 0, 0}

Probably caused by : AuthenticAMD

Followup: MachineOwner
---------

[COLOR=Red]BugCheck 24[/COLOR], {1904fb, fffff8800318c888, fffff8800318c0e0, fffff88001251c25}

Probably caused by : Ntfs.sys ( Ntfs+fc25 )

Followup: MachineOwner
---------

This is a situation to think that something is going wrong there with the hardware.

My first suggestion to you would be re-seat the entire computer once again; making it totally sure that all the component parts are properly seated. Plus make it sure that the two RAM sticks are of the same manufacturer and having the same specs.

Are you overclocking?

CP  F/M/S  Manufacturer     MHz
 2 16,4,2  AuthenticAMD    [COLOR=Red]3013[/COLOR]

As you are getting BSODs, you should stop overclocking and run all the hardware components like CPU, GPU and RAM to their default settings.

Test your RAM modules for possible errors.
How to Test and Diagnose RAM Issues with Memtest86+
Run memtest for at least 8 passes, preferably overnight.

If it start showing errors/red lines, stop testing. A single error is enough to determine that something is going bad there.

Stress test the Graphics Card using Furmark.
Video Card - Stress Test with Furmark

Stress test the CPU.
Hardware - Stress Test With Prime95

Some other information is really needed in the troubleshooting process, but not those texts above. Upload your MSINFO32.nfo file.

1. Click on the start button
2. Type "msinfo32" (without quotes) in the search bar of the start menu, click the resulting link. It will open the System Information window.
3. File>Save. In the "File Name" filed, put "MSINFO32" (without Quote), give the save location to desktop, and click the "save" button.
4. Give the time for processing, it will save a .nfo file on your desktop.
5. Zip it, and upload it following the instruction.

And report us the heat of the computer after a couple of hours of your normal usage. Upload a screenshot of the summery tab of Speccy. Alternatively, you can publish a Speccy snapshot too: http://www.sevenforums.com/tutorials/311593-speccy-publish-snapshot-your-system-specs.html .


Let us know the results.

 

[Carter15]

Thanks for replying,
my personal guess is that there is something wrong with motherboard.

I reseated and cleaned of dust everything because of it about one week ago, didn't help.
Memories are the same for sure. 2x2GB pack, never changed them.
Im NOT overclocking, its some maximum turbo core clock as I found out on net and I never oc'ed before. And I also changed mb battery and reset bios about 2-3 weeks ago, so the oc's would reset.
Yesterday I've ran the memtest for 6 passes (took more than 6 hours, then I had to turn off pc and go to work), no errors. I'll do another memtests tonight with one stick of ram plugged in.
Stress tested GPU in Furmark for a half an hour, max temp 68C, no artifacts.
CPU for about an hour (all tests prime95 is cycling through in small fft mode) max temp was 62C.
No crash or bsod during both stress tests. Funny is I got Kernel-power ID 41 task 63 crash about 5 minutes after exiting prime95, no dump created.

Im also having crashes after POST when Win is booting. Win logo is showed for like 5 secs and then it crashes/bsod's, there is a blue screen shown for like 0,1sec.
Today after the first crash I tried to boot again, asked me if I want to do Startup repair or boot normally, selected to boot. Then it crashed once more, so I selected Startup repair that it offered me. It was scanning for problems about 20 minutes, then it wrote that the problems cannot be automatically repaired and some error codes. Im including the log file it made.
Got one dump from those two startup crashes/bsod's.

MSINFO32, new dump, log and Speccy screens are in zip.

 

[Arc]

Im also having crashes after POST when Win is booting. Win logo is showed for like 5 secs and then it crashes/bsod's, there is a blue screen shown for like 0,1sec.

Before blaming the motherboard (as we cannot test it) better you do some more jobs, IMHO.

The BSODs you are describing might be two other things.

It may be the Corsair headset that is causing the issue. It is somehow old.

    Image name: CAHS164.sys
[COLOR=Red]    Timestamp:        Thu Jun 16 12:40:06 2011[/COLOR]

Does it do the same if you remove the headset? Test please.

If no change, it may be a cold boot BSOD, caused by inappropriate RAM timings. Your timing is 1333 MHz.

  Speed                         1333MHz

Try any other RAM speed? From Motherboards - M4A785TD-M EVO - ASUS .....

4 x DIMM, Max.  16  GB,  DDR3  1800(O.C.)/1600(O.C.)/1333/1066 ECC,Non-ECC,Un-buffered Memory
  Dual Channel  memory architecture
[B]  *AMD AM3 100 and 200 series CPU support up to DDR3 1066MHz [/B]

Can you set the RAM speed to 1066 MHz as a test?

Let us know the results.
_____________________________________

BugCheck 124, {0, fffffa800511b028, b465a000, d8000128}

Probably caused by : AuthenticAMD

Followup: MachineOwner
---------

 

[Carter15]

The CAHS164.sys is a C-Media USB Audio Driver (WDM) developed for Windows. From what I found this is the latest version.
I tried to run pc with just mouse and keyboard plugged in, didn't help, still got crashes.

And with the memories, I searched through bios and I didn't found any other timing settings than those on screenshot I uploaded. The Memclock value ranged from 400 to 800MHz. And the Timing modes are just 4T, 5T, etc. If you know where to set it to 1066MHz, tell me please.
I also think if the problem would be memory timing, I guess it'd show up earlier than after 4 years.

"AMD AM3 100 and 200 series CPU support up to DDR3 1066MHz" has to do something with Opterons, Athlons or Semprons. I searched for it, if it applies for my CPU but noone really knows what that means and what CPU's are involved.

From AMD Phenom? II Processors

[B]Support for[/B] unregistered DIMMs up to PC2 8500(DDR2-1066MHz) and PC3 10600 [B](DDR3-1333MHz) – AM3[/B]

This is for all Phenoms II I guess, I didn't found any memory specifications for my CPU.

One new issue with my pc, I have power on and HDD "work" LEDs on case front panel.
When booting or after crash, in like half of the times the power on led doesn't turn on, fans are running but led is off and pc is not booting. So to boot I have to press repeatedly hard reset button or turn off power on PSU for a while and then try to boot again till the power on led goes on.

 

[essenbe]

The majority of the symptoms you describe sound very much like a Power Supply going bad. You said fans come on but Power on HDD does not come on. It does not take much power to run the fans and a few LEDs, but it takes quite a bit more to power the whole computer.

In the times you reapplied thermal paste, did you remove the CPU?

 

[Carter15]

Fans run and HDD spins too, I said that the pc's power on LED doesn't turn on, when it does then bios shows up. The HDD LED just indicates if it's doing something or not.
Also when we are speaking about PSU, from time to time it's fan starts to run really fast and loud, like if it's cleaning itself from dust automatically (but it doesn't, at least manufacturer doesn't say it has function like that). Duration of it is random, sometimes it runs 15mins, 2mins, few seconds, etc.

Yes I did remove CPU, I cleaned whole pc back then.

And while ago pc crashed again, no blue screen, just crash and pc restarted. 0x124 dump from it.

 

[essenbe]

Hi Carter, I do know what the HDD Led means and how it works, and I would assume it not coming on, means something. I also know that the fan spinning up on your PSU also may mean it is overheating and trying to cool itself off. If I were you I would borrow a PSU from someone and try it for a day or so, just to see.

Failing that I would use some good light and closely inspect the CPU pins to make sure none of them are bent. A bent CPU pin can be disastrous.

I'm sure you are aware that you are getting a 124 error. A 124 error means you have a hardware problem with something in your computer. It is quite a generic code and gives no hint of what the piece of hardware may be. If you followed Arc's excellent instructions, you have tested almost everything in the computer that can be tested. That would leave the problem to be the PSU or the motherboard. The PSU can be tested at idle easily. What is extremely difficult to test is what it does under load, and how long it can sustain that load.

 

[Carter15]

Hi essenbe, simply the HDD led indicates it's activity. The problem is the Power on LED, which should glow when I turn on pc, but sometimes it doesn't, simple as that. Because of it I assume there is problem with motherboard.
PSU spins like that randomly, even when I cold boot (after night when pc was off).

I'll check the CPU, but I was extremely careful when I was manipulating with it.
Im also going to test PSU on multimeter soon and replace it with another one for some time. So we will see.
I have a PSU with about 100W more max power than needed for my build and Im not loading it alot.

And today I got another 0x101 BSOD which stayed showed till I hard reseted pc. Before the BSOD happened, Windows was lagging for about a minute, then bsod. Laggs were like that screen and sound completely freezed for 5 seconds then unlagged for a 2secs then lagged again, etc. Uploading dump.

 

[essenbe]

Carter, before we go any further, I will need an unedited http://www.sevenforums.com/windows-...ne-activation-issue-posting-instructions.html

 

[ICIT2LOL]

Now Carter my 2 cents worth what do you mean by this
"I've read about trying to reseat cpu heatsink"
It is implying to me that you have uncapped the CPU - taken the top metal plate off the CPU and applied compound to the internal chip (much as you do with GPU chips) and redone the thermal compound under the lid of the CPU?? because one does not muck around like that and the compound is usually only applied to the cooler heatsink and the CPU "lid".

Also I read that the BIOS is telling you a new CPU has been installed means you actually took it our of it's socket and reinserted it.

 

[Layback Bear]

This thread is like reading a mystery novel.
I will shut up and go back to watching.

 

[ICIT2LOL]

This thread is like reading a mystery novel.
I will shut up and go back to watching.

Hmm waiting for reply (and chapter 2!!)

 

[Carter15]

Carter, before we go any further, I will need an unedited http://www.sevenforums.com/windows-...ne-activation-issue-posting-instructions.html

I don't really want to post this kind of informations in a public forum. Is it really that needed. I have problems with bsod's, not Win activation.

Now Carter my 2 cents worth what do you mean by this
"I've read about trying to reseat cpu heatsink"
It is implying to me that you have uncapped the CPU - taken the top metal plate off the CPU and applied compound to the internal chip (much as you do with GPU chips) and redone the thermal compound under the lid of the CPU?? because one does not muck around like that and the compound is usually only applied to the cooler heatsink and the CPU "lid".

Also I read that the BIOS is telling you a new CPU has been installed means you actually took it our of it's socket and reinserted it.

What I've meant is reseat cpu's fan heatsink. I didn't remove the metal plate. I reapplied it ON the CPU's lid. And about reseating it, actually enough to remove the bios message was to move with the heatsink like by 1 milimeter or push it a little so the force acting on CPU is spread equally.
Yes I took it off, cleaned it of old paste inserted it in socket and applied new.

 

[essenbe]

Carter, if you will go to BSOD Help and Support - Windows 7 Help Forums , you will see over 1000 pages of different threads, Every one of them posted the information requested. You saw fit to edit the PID of your operating system. There is nothing personally identifiable in that. It only identifies the type OS you are using. The only reason a person would see fit to edit that out is if he were running an illegal system, which is against forum rules. Therefore, before you receive any further help, we will need the unedited http://www.sevenforums.com/windows-...ne-activation-issue-posting-instructions.html. But, I think you know all of that.

 

[ICIT2LOL]

Carter, before we go any further, I will need an unedited http://www.sevenforums.com/windows-...ne-activation-issue-posting-instructions.html

I don't really want to post this kind of informations in a public forum. Is it really that needed. I have problems with bsod's, not Win activation.

Now Carter my 2 cents worth what do you mean by this
"I've read about trying to reseat cpu heatsink"
It is implying to me that you have uncapped the CPU - taken the top metal plate off the CPU and applied compound to the internal chip (much as you do with GPU chips) and redone the thermal compound under the lid of the CPU?? because one does not muck around like that and the compound is usually only applied to the cooler heatsink and the CPU "lid".

Also I read that the BIOS is telling you a new CPU has been installed means you actually took it our of it's socket and reinserted it.

What I've meant is reseat cpu's fan heatsink. I didn't remove the metal plate. I reapplied it ON the CPU's lid. And about reseating it, actually enough to remove the bios message was to move with the heatsink like by 1 milimeter or push it a little so the force acting on CPU is spread equally.
Yes I took it off, cleaned it of old paste inserted it in socket and applied new.

Well that is why you are probably getting the BIOS message - taking the CPU out of the socket. The compound replacing unless you find a real mess from the previous application is best done leaving the CPU in the socket. Now personally I use the Arctic Sliver compound and the cleaning kit when I am doing this and you should never apply more force to the cooler than the manufacturers fixings apply themselves you risk damaging the CPU and the socket if not the board holding them.
The Arctic Silver stuff you can see here Arctic Silver Incorporated - Arctic Silver 5 & the cleaning component/s Arctic Silver Incorporated - ArctiClean at least I find they are the optimum way of cleaning the surfaces many other just wipe or use alcohol and it really is a personal choice.
You only need to apply a small amount of the compound and I use a vinyl glove and my finger to spread the compound real thin on both surfaces and then a small rice grain sized piece of compound on the CPU lid before the final replacement of the cooler. The heat of the CPU operating is enough to ensure the compound is evenly spread as it then it "runs" into those small microscopic imperfections on the CPU lid and the cooler surfaces making an almost perfectly close contact a possible.