BSOD, error 0x000000f4, caused by ntoskrnl.exe

Greetings,

 

I have in the past month been having several BSOD.

 

I have been trying to fix it by changing RAM blocks, re-installing Windows several times, updating firmware of my SSD (Corsair Force GT 120gb), secure erase my hard drives and updating all my drivers.

 

None of it solved the problem. I was thinking that maybe my SSD could cause the problem?

 

My latest crash report is: http://dl.dropbox.com/u/10066760/Crash1.txt

 

How do I solve my problem?

Thanks in advance.

-Try using a dedicated power cable for the SSD

-Try and disable "turn off hard drive after" in Power options

-Try new SATA 6G/B cables

-Use different ports

-If you are overclocked, Save your BIOS settings and reset to defaults.

-Try increasing the PCH Voltage to 1.10v or 1.15v

Try disable TRIM until they fix the drive. You can try those many steps (i wish you luck), or just mine.

That's the typical kernel power loss error by sandforce controller. You're correct in your assumption.

 

Disabling TRIM will not help.

 

Toasted's recommendations may help.

1 Month+ stable without TRIM

 

compared to bsods within 1-2 days with it enabled.

 

if increasing pch voltage helps while having trim enabled without any device failing due to increased voltage, then i would go for this route.

 

but really, i rather just wait for sf to fix their drives.

I have too BSOD 0x000000f4 some times when i left pc alone.

 

But i'm with "higher performance" and "hdd ever active" in my energy saving option :(:

I had the same BSOD but after I uninstalled Comodo Optimizer all problems dissapeared.

Im having the same problem,

here's screenshot of BSOD results, same ntoskrnl.exe error, which how i know caused by disappearing ssd from system. Need power off/on to get back to work!

I think I may of been hit by this today first time for me since replacing the drive a month or so ago.

I cant believe these drives are so fragile that using the wrong sata cable even if its a new one but has a right angled plug on it will cause the drives to crash.

How on earth does that not get picked up in testing before release.

Ape, what are you talking about?

He is talking about unfinished products released to the market thats needs special conditions to work "properly". If properly means BSOD every couples of days instead of hours.

those "special conditions", not work so properly for all ;) but i have lowered BSOD's, may be once at week :P

Seriously. It's a shame that the good reputation of Corsair get soiled by this bug.

 

Sorry for my bad english.

Hi there,

 

I think I'm having the same problems - random BSOD with the following:

 

Bug Check String : CRITICAL_OBJECT_TERMINATION

Bug Check Code : 0x000000f4

Parameter 1 : 00000000`00000003

Parameter 2 : fffffa80`08228b30

Parameter 3 : fffffa80`08228e10

Parameter 4 : fffff800`031e88b0

Caused By Driver : ntoskrnl.exe

Caused By Address : ntoskrnl.exe+7cc40

File Description : NT Kernel & System

Product Name : Microsoft® Windows® Operating System

Company : Microsoft Corporation

File Version : 6.1.7601.17640 (win7sp1_gdr.110622-1506)

Processor : x64

Crash Address : ntoskrnl.exe+7cc40

Stack Address 1 :

Stack Address 2 :

Stack Address 3 :

Computer Name :

Processors Count : 8

Major Version : 15

Minor Version : 7601

Dump File Size : 287,848

 

Windows 7 64bit, Force 3 120gb with v1.3 Firmware and latest BIOS on my motherboard (Asus P6T Deluxe v2). I'm not using RAID but have AHCI enabled and running the drive off the Intel ICH10R SATA controller. When the machine restarts, the BIOS is unable to find the SSD. It requires a complete power reset. I suspect based on the BSOD error message this is some kind of power issue with the SSD?

 

They tend to only be when I leave the computer for an extended period of time - going to try the suggestions above and see if that fixes them (the disabling the power save on hard drive sounds interesting).

 

One thing - I noticed this site:

http://www.sevenforums.com/crashes-debugging/135664-bsod-ntoskrnl-exe-unable-figure-out.html

 

Check to see if your ASACPI.sys is dated 2005 - if so go to this site and download the "ATK0110 driver for WindowsXP/Vista/Win7 32&64-bit" from this site to install a later version (under utilities):

http://support.asus.com/download.aspx?SLanguage=en&m=P6T%20Deluxe%20V2&p=1&s=29&os=30&hashedid=iRlP8RG9han6saZx

 

It reduced my BSOD's considerably - but I still get them from time to time. Hopefully this goes some way into helping Corsair fix the problem as the SSD as has been rock solid otherwise!

This BSOD issue is not corsair's fault. Every SSD on the market with the SandForce 3 Controller in it is having this problem. It's completely rampant. You can read on other manufacturer's forums this exact issue.

 

It's driving my crazy!!!! Only drives with SandForce 3 and it doesn't matter if you are on the x58 z68 or Marvell SATA Controllers it happens to everyone. Sometimes 20 times per day sometimes only once per week. Doesn't matter what your doing on the PC either. It could just be chilling with default BIOS and it will just pop a BSOD.

 

 

Maybe Corsair can work with SandForce and come up with a solution. They would sweep market share if they solved this issue.

some thing worked for me, and now about two weeks it work, well stable - try to disable energy saving for hard drives in power settings (set to "off", in windows 7)

Try Firmware 1.3.2

http://forum.corsair.com/forums/showthread.php?t=99366

Try Firmware 1.3.2

http://forum.corsair.com/forums/showthread.php?t=99366

 

If you runing 1.3, there no need to update to 1.3.2 for fixing temperature parameter, right?

 

Due to NDA there is only 1 change log item available.

 

1. “Fix an issue where SMART attribute 194 erroneously reported 128C instead of 30C when no temperature sensor was present”.

No, it doesn't only fix the temperature parameter. it fixes other things but due to NDA they are not allowed to mention it.

NDA about fixes? lol :) but ok... anyway after disabling power saving features, SSD about two weeks work perfect and stable. Better to leave it as is ;)