Computer shuts down like power outtage, Kernel Power Error ID 41

[Razraal]

Recently I installed a new game on my computer (Far Cry 3), and after a while playing it, my computer suddenly shut down, just like a power outtage (except the power was not out).

After debating with friends, I went more towards a hardware problem rather than a software problem, so I thought something could have too high temperature (CPU, GPU, the HDD, etc.).

I have two HDD's, and they used to be together, but I decided to separate them, to reduce their average temperature.

I installed a software called CPUID HWMonitor, and it shows me these values (when browsing and not playing the refered game):

Note that this has happened before with previous games (Neverwinter and Guild Wars 2), and I'm still not sure whether this is a hardware or software problem, since separating both HDD didn't solve the problem.

Any help is greatly appreciated.

 

[Arc]

Some error relates to Sentinel/SafeNet products.

Event[33197]:
  Log Name: System
  Source: Service Control Manager
  Date: 2013-05-07T09:32:28.187
  Event ID: 7000
  Task: N/A
  Level: Erro
  Opcode: N/A
  Keyword: Clássico
  User: N/A
  User Name: N/A
  Computer: David-PC
  Description: 
O serviço Sentinel64 falhou o arranque devido ao seguinte erro: 
O sistema não consegue localizar o dispositivo especificado.

It is some data protection mechanism, so it may cause some real failure there. And the driver is appearing as too old.

Sentinel64   Sentinel64             Kernel        [COLOR=Red]01-02-2007[/COLOR] 11:39:44

Update Sentinel products from Technical Support - Drivers and Developer Kits.

Also, the just predecessor events of the Event ID: 41 are ....

Event[34173]:
  Log Name: System
  Source: avkmgr
  Date: 2013-05-09T23:33:45.440
  Event ID: 17
  Task: N/A
  Level: Informações
  Opcode: N/A
  Keyword: Clássico
  User: N/A
  User Name: N/A
  Computer: David-PC
  Description: 
[COLOR=Red]avkmgr.sys version 12.0.20.2 successfully loaded[/COLOR]

Event[34174]:
  Log Name: System
  Source: avipbb
  Date: 2013-05-09T23:33:45.549
  Event ID: 17
  Task: N/A
  Level: Informações
  Opcode: N/A
  Keyword: Clássico
  User: N/A
  User Name: N/A
  Computer: David-PC
  Description: 
[COLOR=Red]avipbb.sys version 12.0.50.34 successfully loaded[/COLOR]

Event[34175]:
  Log Name: System
  Source: Microsoft-Windows-Kernel-Power
  Date: 2013-05-09T23:33:45.549
  Event ID: 41
  Task: N/A
  Level: Crítico
  Opcode: Info
  Keyword: N/A
  User: S-1-5-18
  User Name: NT AUTHORITY\SYSTEM
  Computer: David-PC
  Description: 
O sistema foi reiniciado sem primeiro ter sido encerrado de forma correcta. Este erro poderá ser provocado por o sistema ter parado de responder, ter sofrido uma falha ou ter perdido a alimentação de forma inesperada.

Both are avira related. So better you uninstall Avira. At least as a try. Use Microsoft Security Essentials as your antivirus with windows inbuilt firewall, and free MBAM as the on demand scanner.
Download, install and update those, and then run full system scans with both of them, one by one.

Uninstall MagicISO.

ITEM|VALUE
Nome | MagicISO SCSI Host Controller
Fabricante | MagicISO, Inc.
Estado | OK
ID de dispositivo PNP | ROOT\ SCSIADAPTER \0000
Controlador | c:\windows\system32\drivers\mcdbus.sys (2.7.106.519, 249,56 KB (255.552 bytes), 28-07-2012 22:02)

Those SCSI adapters used by the virtual disc emaluator programs are mostly problematic.

Let us know the situation.

 

[Razraal]

I've been using MagicDisc and Avira for a long time, and these problems began occuring only a few months ago, whereas the usage of the softwares mentioned are at least one year old.

I've also previously noticed the Sentinel64 service, and I'm more inclined toward the possibility that it may be the cause.

Nevertheless, I will update the service, uninstall both softwares, install the reccomended anti-virus, and will keep you posted.

PS: This is a trial and error solution, so we must keep in mind that just because the problem doesn't happen in a while ( a reasonable amount of time, say, 1 week or 2) doesn't mean it is completely solved.

Edit: By the way, on the link given to update Sentinel64, which option should be taken on the Drivers and Utilities slider?

 

[Layback Bear]

Do you have all Windows 7 updates including SP-1?

 

[Razraal]

I do.

It just happened again, and again with the same magnitude of randomness.

Edit: All suggestions given by Arc have been applied.
Will post more if any sudden shut down reoccurs.
MalwareBites Anti-Malware encountered and deleted several files (3 identified during scan, more deleted after scan).

 

[Razraal]

After all the suggestions were used, my computer still shut down unexpectedly.
Added further attachment.

 

[Arc]

I've been using MagicDisc and Avira for a long time, and these problems began occuring only a few months ago, whereas the usage of the softwares mentioned are at least one year old.

Lucky you for a long time.

After all the suggestions were used, my computer still shut down unexpectedly.
Added further attachment.

Its difficult to read the logs in an unknown language, but still I am trying. So it needs your cooperation. The zip you uploaded does not contain any crash dump.

Follow it: http://www.sevenforums.com/tutorials/174459-dump-files-configure-windows-create-bsod.html
Go to Option Two, Point 2. Download the .reg file and merge it in registry by double clicking it.

Now wait for another BSOD. When it occurred, search the .dmp files manually in the default path: C:\Windows\Minidump or %SystemRoot%\Minidump. See if the crash dump is recorded or not (hopefully it will be recorded).Post it following the Blue Screen of Death (BSOD) Posting Instructions.

If still there is no crash dump, we may consider it as a hardware issue, and troubleshoot it in that way, but not right now. Observe one or two more crashes and then report us back.

 

[Razraal]

Its difficult to read the logs in an unknown language, but still I am trying. So it needs your cooperation. The zip you uploaded does not contain any crash dump.

Follow it: http://www.sevenforums.com/tutorials/174459-dump-files-configure-windows-create-bsod.html
Go to Option Two, Point 2. Download the .reg file and merge it in registry by double clicking it.

Now wait for another BSOD.

Just to make things clear, none of these sudden shut downs caused a BSOD, so now I don't even know if this thread is in the proper place.
Nevertheless, the file recomended was used.

 

[Arc]

Just to make things clear, none of these sudden shut downs caused a BSOD, so now I don't even know if this thread is in the proper place.
Nevertheless, the file recomended was used.

Yes, a kernel power 41 event is an abnormal shutdown, not necessarily a BSOD. If it is not a BSOD at all, the BSOD analysis will not help there. We need to search for the reason of the shutdown in the hardware at first.

• Check the power cables. Make it sure that it is properly placed.
• Check the PSU, if it is working properly or not. Also let us know the wattage of the PSU.
• Is the computer hot? 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.
• Stress test the CPU.
  Hardware - Stress Test With Prime95
• Are you overclocking? As you are getting BSODs, you should stop overclocking and run all the hardware components like CPU, GPU and RAM to their default settings. Also set the BIOS to default, too.
  How To Clear CMOS (Reset BIOS)

Let us know the results.

 

[Razraal]

• Check the power cables. Make it sure that it is properly placed.
• Check the PSU, if it is working properly or not. Also let us know the wattage of the PSU.
• Is the computer hot? 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.
• Stress test the CPU.
  Hardware - Stress Test With Prime95
• Are you overclocking? As you are getting BSODs, you should stop overclocking and run all the hardware components like CPU, GPU and RAM to their default settings. Also set the BIOS to default, too.
  How To Clear CMOS (Reset BIOS)

Let us know the results.

• The power cables are all properly connected;
• The PSU wattage is 420W;
• Here's the screenshot requested, after one hour of playing;
  
  
  
  
  
• Currently the prime95 stresser is on test 8(approximately 15 minutes total duration so far), will post more as it progresses;
• I'm not overclocking anything;
• BIOS was never changed from its default.

Edit: Prime95 test's duration was approximately 2h, and the last test ran was 800000 Lucas-Lehmer iterations of M-157695 using FFT length 8k. After that, I decided to stop the program.

 

[Arc]

Both the CPU and GPU are running hot.

Dismantle the PC. Clean everything inside. Clean all the fan grills including CPU fans and GPU fans, all the air vents so that it can blow the hot air out easily. While re-seating, apply new thermal paste to the CPU.


Then it is to observe if the temperatures are reduced or not. Post a similar speccy shot when all is done.

A 420 watt PSU is never enough. It will not be able to supply the adequate power that you need. Have a bigger one, at least 750 watts.

 

[Razraal]

So it is hardware, as I suspected.

I guess there is no point updating another SF-Diag zip, but I will do so nonetheless

 

[Arc]

Yes, the diag tool zips are not needed ATM. Apply the suggested changes and then report back with another speccy screenshot.

 

[Razraal]

I have cleaned both CPU and GPU's fans.
Re-applied thermal compound on CPU.
Should I also apply it on GPU? (the same compound)
At the present time I cannot afford a new PSU.

As you can see by the screenshot, CPU's overal temperature suffered a great improvement.
The same cannot be told about GPU (I still need to know the answer to the thermal compound question)

Edit: I'm sorry for taking so long to reply, I've been busy and only had the chance to do it now.
Edit: After correctly applying thermal compound to GPU, the average temperature values, after boot, are as follow:

NOTE that the image previous to the last had a similar temperature on Graphics, however it would always rise to the risk limit of 76-77º C;
now it's stable on 63-64º C temperature.

 

[Arc]

GPU is connected to the PCIe port, no scope to apply the thermal compound there

CPU is showing a better result now. GPU is better, too, but not the best.
It indicates either the GPU is failing to take the load properly, or the power supply is not sufficient (it is true that we already know).

Monitor the temperatures regularly. See if it is going beyond control or not. If it goes beyond control (the letters are showing in red) leave one side of the case open .... point a fan to the direction to blow air there.

Try to replace the PSU as early as possible.

And let us know how it is running now.

 

[Razraal]

GPU is connected to the PCIe port, no scope to apply the thermal compound there

What I meant was applying the compound between the heat sink and the shiny chipset of the GPU, just like on the CPU.

Monitor the temperatures regularly. See if it is going beyond control or not. If it goes beyond control (the letters are showing in red) leave one side of the case open .... point a fan to the direction to blow air there.

I've had one side of the case open in a while (maybe 2 weeks).
Maybe I haven't played long enough, but ever since the cleaning, my computer hasn't shut down unexpectedly.
It seems that the CPU's temp was the one influentiating these shut downs the most.

 

[Arc]

Anyway, replace the PSU as early as possible.

420 Watts is never enough.

 

[mauxie]

imho this problem is ms generated. I used to have this problem, and like other posters,
it only happened during gaming. when I turn my wifi adapter off, I never have this
problem no matter how much gpu power I am using. some will still have a problem
starting games with wifi off, but there is a workaround. open regedit, click on
computer, hit ctrl+f, type in servicelocation, just like I typed it, and wait for the entry
to come up. in the right pane, double click on games, and enter any valid path on
your c drive. this will fake out game explorer, making it possible to start games
without having an active web connection. after you make the change, you may
want to export that part of the hive to somewhere convenient so that you can use it
again after windows changes it back to the original update target.