Vry Strange BSOD - No reason - No Minidump - Please help, desparate :(

[gulliwog]

Hi All

Please find attched all info asked for in sticky, plus as much extra info as I could give, mostly screenshots. I will try to give as much info as I can, without making it too long, so its a little paraphrased.

Background:

This problem started happening yesterday. The computer had already been on for more 24 hours without any problems. No recent hardware changes at all, nothing for months. Regarding software there have been no new programs installed for at least 4-5 days before the problem started, nor made any driver changes for a while. Far as I know there haven't even been any updates. Last Windows update was at least a week ago I think. Skype updated most recently, but wasnt even running at the time of the problems. It seems that from a hardware/software point of view the BSOD comes right "out of the blue", there is no obvious change I can associate with this problem.

Th problem:

The problem itself seems to manifest in exactly the same way each time. After a restart or turning on the PC will work perfectly normally for a minimum of 30-40 mins and up to about a maximum of 60-75 minutes. Somewhere in this time frame whatever I am doing or whatever program I am using will freeze completely. If I am play games the screen will block, if I am online mozilla will white out... or in windows media or photoshop will do the same. Normally at this point I try to use CTRL+ALT+DEL except it doesnt work, and I will invariably get an error saying task manager cant open. After that it is pretty random, sometimes I will get different processes giving errors and stopping, not necessairly the same ones (see images in attachment for examples). Normally a short time after that everything will freeze completely and then the screen will go completely black, followed by the BSOD. A couple of times the PC restarted without the BSOD, but most of the time its there.

The BSOD:

The BSOD itself gives NO INFO about what caused the crash. At the top there is no IRQL_NOT_LESS_OR_EQUAL or any equlivant error, also no minidump file is produced... I have made sure that the minidump is setup so it SHOULD be made. I have tired with both small dump and kernel dump and neither work

Thing I've done:

Ran Memtest86+ v4.20 for 7.5 hours in which time it amde 4 full passes of the 16GB of RAM. It turned up no errors - see attached file for screenshot of end of test.

Removed 2 by 2 the memory cards, running the 2 that stayed in the machine on slots DDR3_1 and DDR3_2. Error persisted.

Use Driver Fusion (new Driver Sweeper) to completely remove all nVidia graphics drivers and reinstalled latest.

Disabled as many processes on startup as I can, not many are running, I try to keep background "noise" as low as possible.

Ran Checkdisk, but I have an SSD as my C:/ drive so I dont know if that really helps. I installed SSD life Free, but I have 2 SSDs in my PC and the free version only gives info on one - Im not sure which one its testing, but it said it was fine.

I have found one or two forums that kinda suggest that a STOP 0x000000F4 can come from a simple CMOS battery problem - given my symptoms, could this be possible?
Stop: 0x000000F4 - Tech Support Forum

One other thing I noticed which may not be relevent, but if the PC sleeps before the BSOD, it can sleep fine for hours, on waking it works like its just been started, fine for 45 -55 mins or so then crash.

I have given as much extra info in the attachment as I can, all the screens from CPU-Z give my system specs. Images of the errors and bluescreen itself. I re-iterate there are NO NIMIDUMP files because they're not getting made. Also info included from a program called RAMMon.

I really hope someone can help me with this, as I have 2 major problems that make it more difficult. The first is I lost my job recently and have literally no income atm, so even buying a new battery is hard decision - especially if Im not sure it will help at all. Secondly the one year guarentee date for most of the things in my PC expires literally in 4 or 5 days... so if its a hardware problem I really need to identify it ASAP.

I would be so grateful for any help, I know the lack of minidump files dont make it easy, if someone can tell me a way I can get it to make the minidump correctly then I will be happy to do it, Im not sure if its something to do with the C:/ drive being an SSD.

Thanks to anyone in advance,

Gulli


P.S. Sorry forgot to say: I scanned with Microsoft Security Essentials, AntiMalwareBytes and Spybot with no resulting problems.

 

[dtxx]

Try system file checker - sfc /scannow

 

[gulliwog]

Thanks, I tired that. At the end of scan I got the message:

Windows Resource Protection did not find any integrity voilations.

I am no expert but I am guessing that means everything is ok?

 

[gulliwog]

Can anyone explain to me how I can diagnose my SSD drive to find out if there are any errors coming from the drive? There are 100s of different programs for HDD health but none for SSD. I have tired SSDLife, but the freeware version doesnt give enough information to be sure.

Please help...

 

[gulliwog]

Update: New Information

The last BSOD I had at the bottom I had a little more info on the minidump not working.

It gave me this error:

Physical Memory Failed with status 0xC0000010

I google this obviously and whats out there seems to be pointing towards the SSD drive, but hardly conclusivly.

I know lots of people need help here but atm I have $2000+ machine that is not suffering occasional problems, its completely unusable.

Oh, and I tried to update the firmware for my crucial SSD, but it keeps crashing and asking if I want to repair windows instead - does that help?

 

[tonyng88]

Hi there, I have got the same problem too.
Got BSOD after windows start up for an hour, with STOP 0xF4 (0x3 xxx xxx xxx), and the line stop at "Initiating disk for crash dump..." .
I still can't figure which part of my computer cause this problems, but probably should be the harddisk problem.
I install my Windows on a SSD drive, then have a 1TB HD and a 500GB HD for saving data.

I am at office now so i can't test and fix my desktop.
Can you try to enter safe mode and see if the bluescreen show up again?

Btw, I have done a lot of researches and see 1 post:
BSOD - STOP 0x000000F4 about 10 mins after startup - Tech Support Forum
He seems to have the same problem as us, but he solved the problem with no reason...

You may also see this:
Hard Drive Diagnostic Procedure
See if the DFT helps or not.

 

[gulliwog]

Hi Tony,

Thank you so much for your respone. Testing to see if the PC crashes in safe mode was next on my list. For the last few hours I have been trying different memory combinations in different sockets. I am on my last try now.. so next time the PC goes down I will reboot in safemode and see what happens.

I read the other thread you linked and there seems to be one major difference between our problems, is that mine crashes within about an hour EVERY time the PC is switched on. I really hoped this time-frame issue would help to diagnose the problem. The PC never crashes straight away - it always works fine for at least 30 mins, usually more. Then SOMETHING changes and programs start freezing 1 by 1 and the crash follows. I kinda hoped that would give someone a clue as to the problem.

I ran RealTemp and the processor temps are fine... I have a very well ventilated ANTEC case with noctural processor cooling. I dont know if its very likely that the memory could overheat, and Memtest did run for 7 hours with no problems.

As for diagnostics, my SSD is made by crucial and there are no diagnostic tools on their site, I downloaded the firmware update but it wont install, instead I get a message about repairing windows. I have downloaded the Seagate SeaTools for windows and ran all the basic tests and everything came back ok. Still no minidumps being produced. I found a couple of threads that said it could be the CMOS battery. I dont know if the fact that no minidumps are being made points to the drive itself.

There is a person that talks in this thread (Juke Chou) about reasons for no minidump, but I dont really understnd it:
Windows 7 x64 always go to Blue Screen in idle STOP 0x000000F4 Physical memory dump with status 0xC0000010 (Fail to dump)

and heres one of the threads that talks about the battery:

STOP 0x000000F4 - Tech Support Forum

Thanks again, Ill post about safemode when I know.

 

[tonyng88]

More info:

I am also using Crucial M4-CT128M4SSD2 SSD. See if the following links help:
http://www.sevenforums.com/crashes-...00000f4-happens-idle-not-no-pattern-seen.html
http://www.sevenforums.com/crashes-...ion-1-hour-after-cold-boot-3.html#post1793840

How long have you brought your SSD? As mentioned in the second link, m4 drive become unresponsive after 5184 hours of power-on time (equals to 216 days of time) and need a firmware update.

 

[karlsnooks]

Gulliwog,

Read carefully. Carry out each and every part of following:

[FONT=&quot]MEMORY DUMPS - REQUIREMENTS[/FONT]


Enable automatic management of the Paging file
WIN + PAUSE/BREAK key combo |
Advanced system settings (left-hand side) |
Advanced tab, Performance section, Settings |
Advanced tab, Virtual Memory section, Change |
checkmark Automatically manage paging file size for all drives |
OK | OK | OK



STARTUP AND RECOVERY SETTINGS
WIN + PAUSE/BREAK key combo |
Advanced system settings (left-hand side) |
Advanced tab, Startup and Recovery section, Settings |
System Failure section | checkmark Write an Event to the system log
| uncheck Automatically restart |


Under Write debugging information, select Small memory dump |
in the Small dump directory: box, enter %SystemRoot%\Minidump| OK

If Overwrite any existing file is checked, then:
Ø Under Write debugging information, select Kernel memory dump
Ø Uncheck Overwrite any existing file
Ø Under Write debugging information, select Small memory dump
Ø OK button
If you get a crazy message from Win 7 about your paging file being too small, then just ignore the message since we have already made sure that the paging file is set to be managed by Win 7.

ENABLE THE WINDOWS ERROR REPORTING SERVICE
WIN key | type SERVICES.MSC | ENTER |
Scroll down to Windows Error Reporting Service |
Rt-clk on Windows Error Reporting Service | Properties |
General tab | for Startup type, choose Automatic |
In Service status, Start button (unless grayed out) | OK


ASSURE THAT CCLEANER IS NOT DELETING DUMPS
CCLEANER | Cleaner icon | Windows tab | System category |
Uncheck Memory Dumps
· WINDOWS ERROR REPORTING SERVICE must NOT be disabled!
· WIN | Services.msc | Enter
· Under Name, locate Windows Error Reporting Service
· If Status is not Started, then
Right-click on Windows Error Reporting Service |choose START
· If Startup Type is not Automatic, then
Right-click on Windows Error Reporting Service |
choose Properties | General tab |
set Startup type to Automatic
· Close Services.msc (I like to use ALT + F4)

After a crash, the minidump file will be in %SystemRoot%\Minidump
%SystemRoot% is generally C:\Windows
========================================
=======================================

After completing above, reboot.

When you get another bsod/crash, check to see if there are any files in:
%systemroot%\minidump

 

[gulliwog]

Hi! Thank you all very much for your answers. Ill start by explain the simplest first:

KarlsNooks: I have already followed all those steps from other posts and stickys. I rechecked everything again and its all setup right except I missed one little thing:

If Overwrite any existing file is checked, then:
Ø Under Write debugging information, select Kernel memory dump
Ø Uncheck Overwrite any existing file
Ø Under Write debugging information, select Small memory dump
Ø OK button

I didnt get this bit right.

I have CCleaner on my PC but so far as I know it only cleans if I tell it to, which I havent. I did uncheck the minidump box anyway.

As regards the SSD and the firmware. I have 2 crucial SSDs on my PC. The first is the C:/ drive and is the older one, bought at the time of the PCs construction almost EXACTLY 1 year ago, its on firmware version 002. The other is newer, from december 2011 and is the F:/ drive used exclusively for games and is running firmware version 009. Now SSDLife Free tells me the 0002 (ie C: crive) has Work Time of 5951 hours (8 months 7 days and 23 hours - powered 1829 times)

Earlier I tired install the windows installation firmware like this and it kept on refusing to install. Instead on restart it jsut gives me a windows repair option before booting normally.

http://edge.crucial.com/firmware/m4/000F/Utility_User_Guide.pdf

I will try the CD version next and see if it works, but heres where it gets a little complicated and I get kinda lost myself. Im not sure if this is relevent at all but when the PC was built the people who put it together didnt bother to change the bios settings before installing windows and the SATA was left in IDE mode. I discovered this after a couple of months while checking something else, at which time I did this to enable AHCI:

GUIDE - Enabling AHCI mode AFTER Windows 7 Installation - Guru3D.com Forums

I double-checked the process to follow on a few sites and did this about 9 months ago. Since then everything has been fine, but could it be this is corrupting the firmware installation now? How should I proceed? Should I put all the SATAs back to IDE (theres three 3 edit in my bios I think) and try again? Or should it work if I use the CD instead of the windows installer? I dont really know that much what I am doing at this level and as there are warnings all over the firmware about messing up drives and data, Im kinda scared to do the wrong thing and no longer be able to start the PC at all. At least I can still use it atm for forums and stuff.

The links about the SSD problem with power on seem very promising, as it looks like my SSD is about the right age for that, but I cant get the damn firmware to install for the moment.

I have to go out for a couple of hours now, I can't get out of it. Ill try these as soon as I get back.

Thanks all again very much - its suprising how alone and helpless you feel when something like this starts happening, especially if your PC is generally more important to you than water, air, food and sex!!

 

[karlsnooks]

First: SSD Life is rather useless. This program and similar programs base their statements on the S.M.A.R.T. values that the SSD reports. Trouble is that no two manufacturers use the same scaling, the same raw values, the same anything. Even a manufacturer will have different ranges of values for different values.

With that said, SIW (link in my signature) will give you a valid report of the data which SIW collects from the SSD.

Yes, I recommend contacting the manufacturer of the SSD as to why you can not update the firmware. I do not know the answer.

Yes. You want to use AHCI.

I need you to go to an elevated command prompt and run following:
wmic recoveros get /all /format:list > %userprofile%\desktop\dmpinfo.txt

this will put dmpinfo.txt on your desktop. Attach that file to your next post.

Here is how to go to an elevated command prompt:
Go To an Elevated Command Prompt


Elevated command prompt = Run CMD.EXE as administrator

WIN | type CMD.EXE | do not hit Enter |
in the PROGRAMS list, right-click on CMD.EXE |
choose "Run as administrator" |
Click on the YES button (if such appears)

WIN key = key with Microsoft log on top

for the guru:
WIN | type CMD.EXE | CTRL+SHIFT+ENTER key combo | ALT+Y keycombo

 

[gulliwog]

Here is the file you asked for... it doesnt actually get put on my desktop, instead I have to goto C:/users/username/desktop but maybe this is because my My documents, my music etc etc has benn moved onto the D:/ drive.

I got another BSOD just while reading your last post... still no minidump

I have to go out now... really really late for meeting. Back in an hour or so.

PS I'll instal SIW when I get back.

 

[karlsnooks]

settings are correct.
You are the 2nd person who moved the Desktop folder to another partition. I can agree with moving the big folders over, such as, documents, videos, music. The others are small. Oh well, makes little difference.

Did you reboot?

BSODs will normally, but not always produce a minidump. Did you look in c:\windows\minidump for .dmp files?

 

[gulliwog]

Sry when I said I moved it over, I didnt mean recently, I did it when I first got the PC. There is still nothing in the minidump file.

I just went by the store where I got it to ask whether I should try installing the firmware or just bring it straight to the and I wound up getting in an arguement with the guy who runs the place. Unfortuneatly I live in the south of france and for those who dont know, customer service here is from the stone age, and he just got bitchy as soon as I tried to ask him anything.

What I want to know is this. If I try installing the firmware again am I seriously risking wiping out everything that is on the C:/ drive? Or is it more one of those 1 in a hundred thousand problems that normally never happen, but are mentioned just in case?

I installed SIW and this is what it gave for the SSDs:

siw.jpg

But is there anything else I am meant to do with this software?

 

[karlsnooks]

Probably nothing disastrous will occur, but, don't blame me if it does and you lose any data dear to your heart. In other words, to be safe then backup any data of importance to you.

Contacting Crucial would not be a bad idea. What do you think?

 

[gulliwog]

Of course I would never blame or hold responsible, its just I am on unfamiliar ground here and am trying to get a feel for things. There isnt really any DATA to speak of on the C:/ drive that I would lose, its all my data is on other drives and/or backed up. Its more that if I crash the C: drive I would have to reinstall everything from scratch, and if the drive is comprimised I dont have any spare drive to install windows on. I havent contacted Crucial yet because I thought simply that they would take days to respond, and as I said with guarentee is up in a few days.

I think I am going to take it back to the store, that way if it messes up its their responsibility. Thing is I just dont really trust them much, they installed my SSDs wrong to being with, and when I spoke to them on the phone yesterday about this problem they said, "oh, its the memory, or the HDD, or corrupt software". Now I agree these are the most common causes for a BSOD, but the story doesnt end there, and if their only going to test the memory and SSD integrity before declaring that the problem isnt hardware, Im worried.

 

[karlsnooks]

What is the precise designation by Crucial of your SSD?

If you need, SPECCY will show you.

 

[gulliwog]

Im not sure what you mean by "designation" below is everything given in speccy for the C: drive

It says it cant determine the manufacturer of either SSD drive. In SIW it recoginsed the C: drive as Crucial, but not the F:/ drive. And it says: 100 read error rate (100 worst) but then at the bottom it say "status good", so Im confused.

The serial number for my SSD is CT128M4SSD2 2.5" SATA 3 C400 (if that helps)



Hard Drives
M4-CT128M4SSD2
Manufacturer Unknown manufacturer
Heads 16
Cylinders 16383
SATA type SATA-III 6.0Gb/s
Device type Fixed
ATA Standard ATA8-ACS
Serial Number 0000000011240311E3A8
LBA Size 48-bit LBA
Power On Count 1831 times
Power On Time 248.0 days
Speed, Expressed in Revolutions Per Minute (rpm) Not used (SSD Drive)
Features S.M.A.R.T., APM, NCQ, TRIM, SSD
Transfer Mode SATA III
Interface SATA
Capacity 119GB
Real size 128,035,676,160 bytes
RAID Type None
S.M.A.R.T
01 Read Error Rate 100 (100 worst) Data 0000000000
05 Reallocated Sectors Count 100 (100) Data 0000000000
09 Power-On Hours (POH) 100 (100) Data 0000001740
0C Device Power Cycle Count 100 (100) Data 0000000727
AA 100 (100) Data 0000000000
AB 100 (100) Data 0000000000
AC 100 (100) Data 0000000000
AD 100 (100) Data 000000001C
AE 100 (100) Data 0000000000
B5 100 (100) Data 0000C800EE
B7 SATA Downshift Error Count 100 (100) Data 0000000000
B8 End-to-End error / IOEDC 100 (100) Data 0000000000
BB Reported Uncorrectable Errors 100 (100) Data 0000000000
BC Command Timeout 100 (100) Data 0000000000
BD High Fly Writes (WDC) 100 (100) Data 0000000051
C3 Hardware ECC Recovered 100 (100) Data 0000000000
C4 Reallocation Event Count 100 (100) Data 0000000000
C5 Current Pending Sector Count 100 (100) Data 0000000000
C6 Uncorrectable Sector Count 100 (100) Data 0000000000
C7 UltraDMA CRC Error Count 100 (100) Data 0000000000
CA Data Address Mark errors 100 (100) Data 0000000000
CE Flying Height 100 (100) Data 0000000000
Status Good
Partition 0
Partition ID Disk #0, Partition #0
Size 100 MB
Partition 1
Partition ID Disk #0, Partition #1
Disk Letter C:
File System NTFS
Volume Serial Number B647809D
Size 119GB
Used Space 62GB (52%)
Free Space 57GB (48%)


Thanks for all your help, btw.

 

[karlsnooks]

Are you ever going to owe me some beer!

gulliwog,
I went to the Crucial site and read up on updating your firmware.

First, let's determine your current firmware version.

[FONT=&quot]Validating the Current Firmware[/FONT]
[FONT=&quot]Revision[/FONT][FONT=&quot][/FONT]

WIN + E | rt-click on the disk icon for the SSD drive |
Properties | Hardware tab |
in the ALL Disk Drives list, highlight your C400 drive |
Properties button | Details tab |
in the Property selection box (probably shows device description)
click on the down arrow at the right end of that box |
Choose Hardware IDs (probably the very next item in the list) |
in the Value box, you will see several entries.
The right-hand end of the entry will show the current firmware version.

I've included a snip from the Crucial manual to show you what you would see if you have revision 000F which is the latest.


View attachment 223180

Do that for both of your SSDs and tell me what firmware version they show.

 

[gulliwog]

Hey man. I promise, if I could Paypal you beer, you'd be in for a pack of the finest Leffe belgium has to offer (or a Duval if your mentally unstable!!)

SO as regards the SSDs I have already checked the current firmware versions of both. My older SSD the C: drive (the important one) is version 0002. The newer one, the F: drive is version 0009.

However (and heres the crunch) something has been bothering me for ages that I never looked into properly cos everything worked and I was scared to mess things up. Even though my 2 SSD are identical they are not always referd to in the same way or treated the same.

Heres what I mean. My mainboard has 8 SATA sockets (0-7). The handbook says 0 and 1 are SATA 6.0GB connections. 2,3,4 and 5 are SATA 3.0 GB connections. 6 and 7 are 2 x Marvell 88SE9172 based chipsets supporting 6.0 Gbs SATA.

Its hooked up like this (its kinda hard to tell cos the numbering is really unclear which is each of the pairs).
My SSD C: drive and my HDD D: drive are connected to sockets 0 and 1. My DVD is connected to 4 or 5 and my second SSD is connected to 6 or 7.

The thing is that during startup on the first screen I get:

IDE channel 4 Master: M4-CT128M4SSD2 0009

On the second creen I get:

Port-00 = Hard Disk, M4-CT128M4SSD2
Port-01 = Hard Disk, WDC (western digitial serial number)
Port-02 = No device
Port-03 = No device
Port-04 = No device
Port-05 = CDROM TSSTcorp etc...

What I dont get is why one of the SSDs is showing up as a IDE device. The 0009 after the name I assume refers to the firmware, so it must be the newer F: drive.

In the bios its the same, the M4-CT128M4SSD2 shows up as the IDE channel 4 master.

Thing is I dont know what the marvell thing means by sockets 6 and 7. Im thinking I should maybe put the second SSD in the place of the WD HDD and then move the HDD to socket 2 or 3 or something.

Its true that its been like this for a long time, but I dont know if it could be relevent now. Its the main reason why I am a little nervous about the firmware installing correctly.

Attached are some images of startup and bios.