Windows 7 ESENT error in event viewer

W

Wipeout

Limp Gawd
Joined
Sep 12, 2002
Messages
271
EDIT: Just as an FYI for anyone chasing a similar sort of issue, this ended up being a memory issue. Have RMAed the memory back to Kingston and everything is running a-OK!

Been chasing a problem since returning home for the holidays and am running out of ideas - I'm having issues with Win7 Pro x64. Computer was rock-solid before leaving for Christmas, and has been very unhappy since I returned two weeks ago. I've already pulled the OS drive and reinstalled to a different HD. Thought it was fixed, but because I've been working from home since beginning of January I've left the machine on 24/7. Rebooted it last night, big mistake. Every time it rebooted, explorer.exe crashed (has encountered an error and needs to close).

Basically things just aren't working right...and the system was rock solid two weeks prior (and off, so no changes happened during that time).

I turned off CEIP Customer Experience Improvement Program (http://www.blogsdna.com/2137/fix-wi...r-update-has-stopped-working-in-windows-7.htm) to get things working this morning because I need to be able to work...it's sad that my VM is more stable than my desktop!

What I'm seeing is somewhat random - occasionally I get video artifacts, but not always. I'll reboot and get the "explorer has encountered an error" message (or other applications), or Aero will be off for no reason I can figure. I've also, since this started (including after reformatting) I've had a hell of a time with Windows Installer packages. By downloading the packages again, I'm able to get it to work, but I can use the same packages without any issue from any other machine (so the packages aren't corrupt).

I've done safe boot, I've done chkdsk, memory tests, etc. I'm at my wit's end, especially since I'd already formatted the computer.

The ESENT error I referenced in the thread title is as follows:

Windows (3480) Windows: The database page read from the file "C:\ProgramData\Microsoft\Search\Data\Applications\Windows\Windows.edb" at offset 98304 (0x0000000000018000) (database page 2 (0x2)) for 32768 (0x00008000) bytes failed verification. Bit 196065 was corrupted and has been corrected. This problem is likely due to faulty hardware and may continue. Transient failures such as these can be a precursor to a catastrophic failure in the storage subsystem containing this file. Please contact your hardware vendor for further assistance diagnosing the problem.
Click to expand...
Mobo is Asus P5QL Pro
Processor is a C2Quad
Graphics are Sapphire Radeon HD 4800
8GB of Kingston HyperX T1 Series DDR2

Everything is SATA (HDs, DVD burner) in AHCI...could AHCI be my issue? Using latest Intel drivers (same as before).

Is there anything I should be looking at here that might be my root cause?
Don't have spare memory
 
Last edited:
was the computer plugged in, but turned off? on a UPS? it's possible a power surge may have hurt something.... I've had a power flicker kill an on-board NIC before, and a power supply.

what memory tester did you use?

what you may also try is downloading a LiveCD, and run off that for a few hours just to see if the issue persists. If it does, then it would definitely point to a hardware issue.

Wipeout said:
I've also, since this started (including after reformatting) I've had a hell of a time with Windows Installer packages. By downloading the packages again, I'm able to get it to work, but I can use the same packages without any issue from any other machine (so the packages aren't corrupt).
Click to expand...

this sounds like it could be a memory issue
 
Computer was plugged into my UPS, and turned off.

Used MemTest (http://hcidesign.com/memtest/) to test memory for most of Saturday with no issues there. Suppose I could've run it longer than I did (i.e. the whole weekend), but I don't think that would make much of a difference.

Can I run a LiveCD of 7/XP? I have a LiveCD (okay, USB disk) of Ubuntu 9.10 but don't know Ubuntu's possible reactions to problems like I know Windows.

I agree with the Windows Installer issues possibly being memory - other than testing the memory with software, my only option is to start systematically pulling one stick at a time and seeing what happens, right? Or borrow known working memory from a friend and see if issue remains...
 
may want to try memtest86+, as it's the one most often recommended.
http://www.memtest.org/

although the one you used, never heard of it myself, but may be just as good.

and you could try building a BartPE disc, or create a boot disc via UBCD4Win http://www.ubcd4win.com/ since that will get you into the Windows environment (although I don't know if it can use Win7? haven't dinked with it too much myself)
 
I must be losing my mind - I forgot; ran memtest86+ as well this weekend, with no issues reported there either. I'll look into building a PE disk and taking a look at that.
 
Perhaps I will check my video drivers, seeing how my screens just went black and came back up with "Display driver atikmdag stopped responding and has successfully recovered."
 
hard to say....

it just sounds like a memory or possibly mobo issue.
but you've checked the memory.

artifacting on the screen is normally the video card, possibility of memory.

maybe try pulling all but 1 stick of RAM?

edit: anything wonky in event logs?
 
Other than the ESENT error way up there in the first post, just "Faulting application Explorer.EXE" when Explorer crashed this morning (kernel32.dll, by the way).

I guess mobo is a possibility, as is one bad stick of ram (which always sucks since it's a total crapshoot as to which one it is). I was gonna reboot it here now that I'm done working and see where I stand (i.e. any new event viewer errors).
 
doubtful it will do any good, but can always try running an sfc

Start -> Run -> sfc /scannow

make sure no Windows system files got borked somehow :confused:
 
SFC found corrupted files it could not fix. Fawk.

Can you even do a repair install that doesn't bork everything on the machine? And besides, this is a two week old install...I'm not real confident that'll even fix the issue. I've got a system image, but that doesn't do anything if it's got problems too.

j-sta, I really appreciate the assistance thus far.
 
Last edited:
Well, one stick of memory out (was 3rd stick) and things be better. Still have ESENT errors after reboot, and some registry hive issues, but wondering if that's because when I reinstalled I had a bad stick of RAM in.

As long as it stays on I should be okay...really hesitant to reformat/reinstall this weekend on off chance it doesn't work!
 
ok... well since SFC found a corrupt file, that's good and bad. except for that it couldn't replace it.

maybe now that you're down to 1 stick, try sfc again?

although, it's the general consesus that if an OS is installed on a system with bad memory, the best way to cure any irregularities after the RAM is replaced, is to reinstall.

Since who knows what all kinds of issues it could cause.

But, I would recommend trying sfc with the 1 stick, and hopefully it should be able to replace the file.
 
I agree...sure it works now, but what other crap is gonna break in a month, six months, whatever. The fact that the system is recovering registry hive files on reboot makes me more than a little worried, to boot.

I'm going to have to try my luck with Kingston to RMA this stick, because I'd like a working stick of memory. In the meantime, I'll hobble along until the weekend and probably format/reinstall then.
 
Last edited:
Well, it was going to wait until the weekend, then it froze on me right at the end of my work day, so I powered down, disconnected all drives except DVD and the drive I wanted the OS on (which I'd done a full format on), and put in my Win7 disk. This is still with 3 sticks of memory in.

Started to install, then got an error at 0% Extracting files - this is the same problem I had previously (2 weeks ago) that seemed to be fixed by repeating over and over and over again until finally it took. My searching of teh interwebs made me think that was memory-related (like with MSI installer issues), so I pulled all but one stick of memory out and tried again, and install went smooth.

Okay, Windows installed - next hurdle, Office 2007 DVD kept yelling at me last time that the disk was corrupt (Windows installer issues, likely; and bullshit because I've used the disk on other machines, including after the desktop told me that the disk was corrupt!). That also installed successfully.

Tried installing VirtualBox 3.1.0 and get an error that the "product.cab" file is missing or invalid (MSI installer again). Way around that last time was to just re-download file and it works...however I tried installing the same install package that I had on my laptop, works without issue. Same thing with Trillian 4.0.19.

Decided to run MemTest overnight, computer was off this morning when I woke up. Looks like it Bluescreened after I went to bed.

At this point, I can't figure out of the rest of my memory is also suspect or those files (some of which were downloaded on the system that was having lots of errors) really are corrupt. I'm thinking it might be easier to just swap in some memory from a friend that should be known working and see what happens.

...or am I overthinking here?!
 
definitely sounds like memory issue, and it is very possible that the files that were downloaded with bad memory installed, could have been corrupted.

and I don't think Memtest can "blue screen" but I've also never seen it cause a computer to reboot. All I've ever seen it do is freeze the whole system.

run memtest with 1 stick? or were you running it with all 3?
 
if you have access to some known good memory to swap out for a bit, absolutely, thats a good step to take in this troubleshooting process
 
One stick only - I'm at 2GB currently (system had 6 last night when I tried installing Windows the first time).

...perhaps I need to find an installer that was downloaded in December (before the initial issue) and run that?
 
Ok. I've spent most of Saturday running mentest in different configurations. I tested each of my 4 2GB sticks separately. I tested each stick in each slot, then moved to next stick, etc.

I'm borrowing two sticks from a friend - when I put the sticks into their corresponding colored slot (so I was in dual channel), the memtest froze at about 11% for several hours. Could the issue be the memory controller and not the memory itself? I don't see a way of turning off dual channel mode in the bios, so as soon as I go to more than two sticks this happens. This being the issue actually makes sense, seeing how when I started installing Windows again on Thursday, I was in dual channel mode.

What has me worried is what might happen should I test within Windows. After the hosing of registry hives and other things when I installed Windows with this problem previously (and memtest not catching it at that point), I worry about going with dual channel mode. Is there any other way I can test without hosing anything? Guess I could make a system image and then let 'er rip...
 
Think it's solved. I used VirtualBox as my test within Windows (after making a system image in case I broke shit).

With known working stick (#1), can install without issue.
Add stick #2, VirtualBox installer gives MSI errors.
Remove stick #2, replace with stick #3. Explorer.exe crashes. That's probably the corrupted stick.
Remove stick #3, replace with stick #4. VirtualBox installer gives MSI errors.
Remove stick #4.

Put in two 1GB sticks loaned from friend. Install VirtualBox: no errors. Methinks I have one really bad stick and two suspect sticks. Anything I may have missed?
 
I'm sure it's not a very common issue, but once I did come across a system that had issues when memory was running in dual-channel.

I would put in 2 sticks in corresponding spots for dual-channel, and had nothing but issues.
moved 1 of those sticks to the other channel, so no longer running in dual-channel, worked fine.

had 4 sticks total, did lots of swapping and moving around, and determined the issue only occured when memory was in dual-channel.

called up Dell, told 'em the mobo was bad, got the new one and all problems were gone.
 
Well, right now I have a 2gb stick in A1, A2 empty, and two 1gb sticks in the B channel and appear to be OK. So thinking it may not be the channels. Will run it this way for a few days and see what happens, while probably contacting Kingston to get them to replace the memory.
 
You must log in or register to reply here.
Back
Top