Windows 7 Build crashing

[master7045]

Hi all, I have a windows 7 build running on a ThinkServer TS series box (newegg special from a few years ago) that I run headless and just use RDP to connect to. It's my plex server and Blu Ray ripping machine. Recently I've noticed 2 things, one the Blu Ray drive is not reading disks anywhere near the 100% it used to (separate issue) and more importantly, I'm getting a "windows has recovered from an unexpected shutdown" The error message is as follows:

Problem signature:
Problem Event Name: BlueScreen
OS Version: 6.1.7601.2.1.0.256.48
Locale ID: 1033

Additional information about the problem:
BCCode: 7a
BCP1: FFFFF6FC40008118
BCP2: FFFFFFFFC0000185
BCP3: 0000000030621860
BCP4: FFFFF8800102393C
OS Version: 6_1_7601
Service Pack: 1_0
Product: 256_1


Files that help describe the problem:
C:\Windows\Minidump\030916-33821-01.dmp
C:\Users\Matt\AppData\Local\Temp\WER-227168-0.sysdata.xml

I've attached dropbox links for the dmp file and xml as well. Can anyone read the dump file and give some advice?

I tried to do a bit of googling and came across a topic where the users SSD had a bad sector. That made me think about how my backup drive (B is full and I know that windows scheduled backups were failing. I've deleted all old backups and turned off the weekly scheduled backup to see if that clears things up.

Please help!

 

[Bubbaleone]

WhoCrashed analysis of crash dump file: C:\Minidump\030916-33821-01.dmp:

This was probably caused by the following module: ataport.sys (ataport!ChannelQueryDeviceRelations+0x0)
Bugcheck code: 0x7A (0xFFFFF6FC40008118, 0xFFFFFFFFC0000185, 0x30621860, 0xFFFFF8800102393C)
Error: KERNEL_DATA_INPAGE_ERROR
file path: C:\Windows\system32\drivers\ataport.sys
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: ATAPI Driver Extension
Bug check description: This bug check indicates that the requested page of kernel data from the paging file could not be read into memory.
The crash took place in a standard Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system that cannot be identified at this time.

Description from Microsoft hardware dev center: ATA Port Driver

Before you do anything else, run chkdsk. I've seen this error before, and while running SFC will tell you that there was no problems with any system files, you should try running chkdsk following this procedure:

1. Boot your Windows 7 installation disk or USB flash drive.
2. At the first screen click OK which then opens the "Install Windows" screen (don't click any buttons).
3. Press Shift+F10 to open an elevated command prompt.
4. Type the following command, then press Enter:

>chkdsk C: /x /v /r /b

Let chkdsk run to completion without any interruption whatsoever. When it finishes you'll be able to read (and note) a report of any errors that were detected and repaired.

.

 

[Elixer]

No, never run chkdsk until you are certain the rest of the system is OK.

Unexpected shutdown can be PSU, heat, or other hardware issues.
Get CrystalDiskInfo, and run that as well, that will tell you status of your SSD/HDs.
First, check the voltages of your PSU with something like CPUID HWMonitor.
If those are OK, run occt or prime95 with CPUID HWMonitor running, and keep an eye on temps.
If those are still OK, next step is run memtet86+ overnight to make sure RAM is OK.

Hopefully you will find the issue with the things mentioned above. If not, report back.

 

[Bubbaleone]

No, never run chkdsk until you are certain the rest of the system is OK.

Unexpected shutdown can be PSU, heat, or other hardware issues.
Get CrystalDiskInfo, and run that as well, that will tell you status of your SSD/HDs.
First, check the voltages of your PSU with something like CPUID HWMonitor.
If those are OK, run occt or prime95 with CPUID HWMonitor running, and keep an eye on temps.
If those are still OK, next step is run memtet86+ overnight to make sure RAM is OK.

Hopefully you will find the issue with the things mentioned above. If not, report back.

This is the problem as stated by the OP:

I'm getting a "windows has recovered from an unexpected shutdown"...

These are the pertinent pieces from the crash dump findings:

*Error: KERNEL_DATA_INPAGE_ERROR
*file path: C:Windows\system32\drivers\ataport.sys
*Bug check description: This bug check indicates that the requested page of kernel data from the paging file could not be read into memory.

Here, MSDN states the reasons for kernel_data_inpage_error in order of likelihood

This Stop message indicates that the requested page of kernel data from the paging file could not be read into memory. This Stop message is usually caused by a bad block (sector) in a paging file, a virus, a disk controller error, or failing RAM. In rare cases, it is caused when nonpaged pool resources run out. It is also caused by defective hardware.

And the important part, as stated by MSDN, is that "This Stop message is usually caused by a bad block (sector) in a paging file...". The data indicates a high probability that this is a disk error.

Bug checks are one of the most important tools we have to quickly and accurately determine an efficient course of action. And the fastest, safest, most efficient method of eliminating bad sectors as the cause for the shutdowns is to run chkdsk with Widows offline, as I described. If chkdsk comes back with no errors reported the OP is golden and, if the problem persists, should then turn to the excellent diagnostic next-steps you provided him with.

I've worked in this industry for over thirty years and have witnessed all the FUD about chkdsk that's out there.You are, in effect, telling the OP that chkdsk is somehow unsafe. As you seem so emphatic about it, and I have always been very open to learning about anything new, I'd be extremely interested in reading the industry information you base your opinion on. Because, industry best practice in this scenario has always dictated eliminating any file system and disk errors as the very first step.

.

 

[Elixer]

I've worked in this industry for over thirty years and have witnessed all the FUD about chkdsk that's out there.You are, in effect, telling the OP that chkdsk is somehow unsafe. As you seem so emphatic about it, and I have always been very open to learning about anything new, I'd be extremely interested in reading the industry information you base your opinion on. Because, industry best practice in this scenario has always dictated eliminating any file system and disk errors as the very first step.

I am only saying it is unsafe since they need to rule other things out first.
Running Crystaldiskinfo should show the SMART status of said drive, and if indeed that shows the drive is in a cautioned state, then yeah, replace the drive, and continue on.

If SMART shows all is well, then, that is where I have read/seen in many forums/actual experiences where the 'Windows has recovered from an unexpected shutdown' has been caused by drivers, PSU, and the other stuff I mentioned. I have personally seen chkdsk cause more issues resulting in having a even more fubar'ed system, since the problem turns out to be RAM related, or O/Cing too high or things of that nature.
This is why I tend to strongly caution people never to run chkdsk at the first sign of trouble.

 

[VirtualLarry]

Yeah, running CHKDSK at the first sign of problems, on a system with potentially unstable hardware (CPU / RAM / storage), is a good way to kiss the user's data goodbyte.

Edit: Maybe I'm confusing CHKDSK (readonly) with CHKDSK (fix problems).

 

[Bubbaleone]

@ Elixer: Your intent is very good but I believe your methodology is flawed because you bought into all the FUD about chkdsk. I can appreciate how even an enthusiast like you might be uncomfortable using a CUI instead of a GUI to perform critical repair work on an ailing system, especially if you came on the scene after the days of DOS when everything was done via commandline.

The first tool in ones toolbox should be chkdsk. One needs to earnestly study on it's correct usage and then practice using it regularly in order to get over the fear that one's going to destroy their computer by using it. It's an excellent diagnostic and repair tool, not a bomb.

There's a long standing axiom whose acronym is GIGO, meaning "garbage in...garbage out". If the storage device has errors in it's volume bitmap, bad sectors, orphaned indexes, etc., etc., it won't matter what kind of GUI diagnostic tools you throw at it since the data they'll be reporting back is unreliable, because the condition of all the data on the disk is unreliable.

Here's the problem; all the data from the OP's crash dump indicates a serious disk error rather than a hardware problem. Viruses, memory modules, hardware...they're all possibilities as well. And they always are, trust me. But a good admin knows he can't waste a client's data, time, or money else he'll be looking for a new job. So we focus on using time tested and proven methodologies and tools to get that machine back online and fully functional in the shortest, most cost effective amount of time. Running chkdsk to ensure the disks data is as reliable as possible is always the first step in the process of eliminating all the potential causes accurately and reliably. Just remember, GIGO!

.

 

[master7045]

Thank you all for the feedback and discussion. It's always interesting to read options from others and helps me learn in the process. To hopefully clear things up, let me share a bit about this system. I have 4 different HDD on this machine and the Blu Ray Drive. All are 3.5" HDD with their own separate drive letter. C:\ (250GB; OS & programs), B:\ (250GB; Backups), M:\ & E:\ (2TB & 4TB, respectively); contain "documents", which means, it's my one drive folder, local documents, music, and my MKV rips.
I've run CheckDisk on B, M, & E as I could do so via the GUI. B results found errors and were fixed (nothing else was mentioned or location of results). M & E no errors were found. I think I'm going to wait 24-48 hours to see if it crashes still and then will run C: through checkdisk as well as the mem test. i have PassMark Performance Test & CPUID installed already and haven't noticed anything out of the ordinary in regards to fan speeds or temps. That is part of the reason I abandoned a home built PC and went w/ the IBM pre-built, to avoid some of these issues.

 

[ratjacket]

I have never had a problem with chkdsk and I have run it on hundreds on machines

a quick run in cmd prompt from windows for a start then the more intense operation listed above

my only regret is I haven't run it enough on some machines I have wasted time on a clearly failing HDD when I should of known better.

 

[nerp]

Never heard of chkdsk destroying data. I have heard of it being the last thing used before a drive on the verge of dying finally dies.

 

[CorsairDemon]

Yup, it will.

Have you checked in Device Manager and enabled hidden devices? There could be drivers loading that are no longer needed, or are incorrect for your system.

 

[master7045]

Will check for old/bad drivers before running checkdisk. The machine rebooted again yesterday somstillmnot fixed. I actually just bought an SSD to replace the os drive anyway. The HDD has got to be pushing 6-10 years old anyway

 

[RLGL]

Give it up..... update the machine and be done with it