upgrade SATA drives on MS-1719 / GX700

X

XyKy

New member
PRIVATE FIRST CLASS
Joined
Oct 15, 2011
Messages
6
Hello, trying to preemptively replace/upgrade the HDD of my 2008 MS-1719 (Core 2 Duo, 8600M GT) laptop. I made the mistake of replacing my original drive (Seagate Momentus #st9160821as, 160GB, 5400RPM) with a SATA II drive (Western Digital Scorpio Blue 500 GB, 5400RPM, also with a Scorpio Black 500GB, 7200RPM). Long story short, I get a BSOD if I don't reboot every 24 hours! I believe it's due to SATA II not being compatible with my machine.

https://forum-en.msi.com/index.php?topic=152513.msg1128891#msg1128891
Did you install a jumper to limit it to 1.5?
Click to expand...
...no jumpers on these 2.5" drives  :(

I now realize my error and want to install a compatible drive - so my question is: is there a capacity limit for compatibility with this machine? Does anyone know which drives were originally offered with with the 1719?  :hat tip:
 
Well I could be wrong about the reason for my BSOD problem - I won't bore you with the details - but I'm still curious about HDD compatibility.
 
I have had a 1719 running with a SATA II for a couple of years.  I wouldn't think that it is a compatibility problem.  What is the exception code you are getting when it blue screens?
 
Freshy, I appreciate your offer of help.
I don't know - it's complicated. Just walk away now!!
(actually, writing this out made me see something I haven't tried yet - skip down)

Here's my sad story:
- I replace the original drive with the Scorpio Blue; so far so good.
  (cloned old disk with Acronis)
- A week later, I get an ExpressCard USB 3.0 adapter and external drive enclosure.
- I install "MSU Quality Measurement Tool" and and some codecs - that is all.
- experimenting with Fraps writing to the new external drive.
- start getting blue screens
      Error code 10000050, PAGE_FAULT_IN_NON_PAGED_AREA

per http://support.microsoft.com/?kbid=183169&sd=RMVP
  * usually occurs after the installation of faulty hardware or in the event of
      failure of installed hardware (usually related to defective RAM, be it main
      memory, L2 RAM cache, or video RAM)

- Checking event viewer, among the common errors were
    Srv: "The server was unable to allocate from the system nonpaged pool because the pool was empty."
    ftdisk: "The system failed to flush data to the transaction log. Corruption may occur."

- ran WD Align Tool
- ran chkdsk - no problems
- ran Windows Memory Diagnostic for 6 hours - no problems
- ran Spinrite for 6 hours - no problems
- yes I'm impatient & should have run these tests longer

- after a while find I'm OK if I reboot every 24 hours (not 25, 24).
- I uninstall "MSU Quality Measurement Tool"
- exchanged new internal drive with a Scorpio Black, which was my external drive.
  (nice performance boost BTW)
- two months go by, rebooting daily.
- swapped disks again, back to the original equipment - problem persists.
- get fed up, post to MSI forum.

Yes, the BSOD occurs whether the ExpressCard USB 3.0 adapter is inserted or not.

I guess I need to uninstall the adapter, and do without a high-speed external hard drive.
I hate to do it, but it's the only thing left.
If that doesn't fix the problem, it might be an unrelated hardware failure (RAM?)
;-(( :tinyviolin:

I'll update again in 24-48 hours and let you know if that fixed the problem.

...if I can get it working, I'll donate it to family and get a modern laptop.
:)

 
While this could be a hardware failure, I think it is more than likely to be a driver issue.  Fortunately, there are tools available that will help us find the offending component.  When Windows blue screens, it will generally save a mini-dump file.  WinDbg from the debugging tools can analyze the dump file and pinpoint which file was loaded at the address that caused the crash.  Sometimes, this file will be part of the driver that is the source of the problem.  More often than not though, the file will be a Windows file.  When this is the case, WinDbg can show the call stack, looking back through this will allow us to try and find which driver was calling into windows.  It is slightly involved, but not risky.

Go to http://msdn.microsoft.com/en-us/windows/hardware/gg463009.aspx and scroll down to "Debugging Tools: More Information and previous Versions".  Choose the link appropriate for your OS (32 or 64 bit), download and install it.

Then, right click on the "Computer" entry on your start menu.  Select properties, and click the advanced settings link on the left hand side of the window.  Down towards the bottom will be a button labeled "Environment Variables", click it.  In the environment variables window, click "New..." down under system variables.  Enter the following:

Name: _NT_SYMBOL_PATH
Value: symsrv*symsrv.dll*C:\symbols*http://msdl.microsoft.com/download/symbols

Click OK to close this window, and then click OK again to close the system properties window.

The configuration part is over, we can now begin investigating the dump.  On your start menu, find the Debugging Tools folder, and select the WinDbg icon.  In WinDbg, open the file menu and choose the "Open Crash Dump..." entry.  Browse to one of the dump files created when Windows crashes and open it.  A new window will open and spew lots of stuff at you.  Don't worry, you don't need to understand it all.  Mostly, you want to make sure that most of the files that it rattles off get their symbols loaded.  There will probably be a few that don't, and that is OK.  If the symbol files are getting downloaded correctly for most of the files, we can proceed (depending on your Internet speed, this can take a few minutes, also once finished, you may delete the C:\symbols folder).

You will notice down towards the bottom of the window there is an area where you can type in a command.  Type in:

!analyze /v[Enter]

Again, WinDbg is going to spew a large volume of text.  You don't have to understand all of its output.  At this point we are just looking for a single line that identifies the file  that is the probable cause of the error.  If we are lucky, it will be a file in the driver itself.  Otherwise, it will be a Windows file and we will have to look at the call stack to pinpoint the driver.

Feel free to copy the WinDbg output into a text file and attach it to a post.  I will have a look at it and see what I can tell you.  There should not be any sensitive information in it, mostly memory locations and names of system files.

If you do not feel comfortable doing all of this, you can upload the mini-dump file to a file sharing site and I can analyze it for you.

HTH,
Freshy
 
Freshy - thanks again for helping.
Sorry I didn't reply - I started to wonder if I had cloned my drive incorrectly, and have been on the Acronis support forum to inquire.

I think that's been eliminated for the moment. Taking advice from the good folks over there, I'll be running MEMTEST all night tonight to check again for possible bad RAM.

It has struck me today that maybe the external USB 3.0 adapter & enclosure combo has somehow corrupted the disk drivers of my system. The ExpressCard adapter & its driver is gone, but uninstall might not be complete. So, if MEMTEST doesn't find anything, I will probably do a full reinstall of Windows. I'll keep you informed.
:undecided:

OK MiniDumps: here's a typical one, with "gobbledygook" removed:

IMAGE_NAME:  hardware_disk
PROCESS_NAME:  csrss.exe
ExceptionCode: c0000006 (In-page I/O error)
Inpage operation failed at 75b7b399, due to I/O error c000009a


Here are some recurring Event Log messages:

Event Type: Error
Event Source:   sr
Description: The System Restore filter encountered the unexpected error '0xC000009A' while processing the file '' on the volume 'HarddiskVolume1'.  It has stopped monitoring the volume.

Event Type: Warning
Event Source:   Ftdisk
Description: The system failed to flush data to the transaction log. Corruption may occur.

Event Source:    Srv
Description: The server was unable to allocate from the system nonpaged pool because the pool was empty.
 
The first error roughly translates into:

Code: 
The volume label you entered exceeds the label character limit of the target file system.

This is a bit misleading as the text of the event log entry indicates that the volume label is empty (or corrupted).  One of things you can try:

From a command prompt:
Code: 
chkdsk c: /f
Reboot to let this run.

This will fix any kind of corruption in the file system that may have occurred during the cloning.

As for the last event log entry, what AntiVirus are you using?  Almost every time I have seen an exhausted non-paged pool, the AntiVirus has been the culprit.  Device drivers are quite capable of causing this, and given the problems you have been having, I would suspect them, but, lets eliminate your AV as the source first.

There are a couple of AntiVirus solutions out there that are free and will keep you fairly safe.  Microsoft Security Essentials or AVG or something along those lines.  What we want to do is remove your existing one, and install a different one.  If the problem goes away, then we are closer to finding its cause, if not, then we can focus on the drivers.
 
just ram MEMTEST86 all night with no errors...

File corruption eh? Interesting
BTW my "HarddiskVolume1" is a temp partition, with System Restore manually disabled.
"chkdsk /f" has been done since the problem occurred, but I will do it again.

Planning on fresh Windows install at this point.

Re antivirus: I don't run it and haven't for years.  :shocked_anim:
Former AVG user until the bloat and nagging became too much.
Never got a virus in 10 years anyway. I surf safely (famous last words?)
I do run Spybot Search & Destroy occasionally - I figure any virus will immediately add spyware.

EDIT - I run a webserver from my bedroom and that DOES use AVG so I'm not an AVG hater.



 
Well the problem has been resolved so I guess I should update.

1)
I should have cloned my hard drive with the target drive in place & the source drive connected via USB,
not the other way round. This is called "reverse cloning" - see http://kb.acronis.com/content/2931
This was the source of my disk hardware related errors.

I was able to recover from this error by creating an ASR (Automated System Recovery) backup, then
reformatting my hard drive, repairing the MBR, then performing an ASR restore.

After fixing my drive this way, I successfully fixed its' twin (backup) drive by the reverse clone method.

2)
I had also done something to my USB controllers - part of the Intel 965 chipset. I am still not sure
what happened, but unless I disable five of the seven available USB controllers, I get a memory leak
of "pool nonpaged bytes"; the total pool nonpaged bytes slowly and steadily increases until it hits the
32-bit limit of 256 MB, at which point random failures start to occur, shortly followed by a BSoD.

So - if you have a BSoD's occurring on a regular schedule, suspect a leak of "pool nonpaged bytes".
I found it with Windows Performance Monitor, but you can simply use Process Explorer: System Information:
Memory: Kernel Memory (K): Nonpaged. With XP 32-bit in normal boot (without the /3GB switch) the limit
is 262,144 - anything over half that is unusual and should be looked into. The limit is much higher in 64-bit.

Hopefully this will help someone with a similar problem.
 
You must log in or register to reply.
Back
Top