Explorer invalid page fault in and in Kernel32

[dadioh]

legg wrote:

> This left only the explorer issue, which persisted in the absence of
> any system stress, but still triggered by user window manipulation.

Excuse me butting in but have you tried using the Internet Explorer Repair
Tool? I realize your problem isn't with IE but it and Explorer are quite
intertwined and I've had success in flogging explorer into behaving by using
the IE repair tool. It is available via MS System Information>Tools or
Control Panel>Add-remove programs>Internet Explorer>Remove.

dadiOH

[legg]

On Sun, 17 May 2009 06:54:09 -0400, "dadiOH" <dadiOH@invalid.com>
wrote:

>legg wrote:
>
>> This left only the explorer issue, which persisted in the absence of
>> any system stress, but still triggered by user window manipulation.
>
>Excuse me butting in but have you tried using the Internet Explorer Repair
>Tool? I realize your problem isn't with IE but it and Explorer are quite
>intertwined and I've had success in flogging explorer into behaving by using
>the IE repair tool. It is available via MS System Information>Tools or
>Control Panel>Add-remove programs>Internet Explorer>Remove.
>
>dadiOH
>

I tried that finally this AM, after getting two Internet Explorer
error messages for the first time in some years, even though the
program wasn't running. It had been run to access a sticky website
earlier in the morning. These errors occurred consecutively with two
instances of the more familiar explorer.exe error, while transfering
file and opening windows. Other programs running at the same time,
were unaffected, as usual.

The repair option was unsuccessfull.
"Please run setup again to reinstall all component"
Details:
"Version 3.0.8168.0 of file atl.dll exists, but the version needs to
be greater than 3.0.8449.0"

RL

[legg]

>On Sun, 17 May 2009 03:59:10 -0400, MEB <meb-not-here@hotmail.com> wrote:

>>legg wrote:
>> On Sat, 16 May 2009 22:40:07 -0400, MEB <meb-not-here@hotmail.com>
>> wrote:
>>
>large snip
<bigger snip>
> Well, we finally got to the registry modifications. Might have helped
>to advise of that from the start...
>
> So these were done manually?
>
Note that this was done over a year ago. It's just an example of the
methods used, and the records kept.

Invalid keys were deleted through the Norton "WinDoctor" GUI. These
deletions are supposedly 'undo'able, though I've never reverted in
this manner - prefering to restore from registry back-up when needed
and to try again (or not).

Missing dll and excutable files, if suitable ones be found, were
relocated (copied) manually to correspond to the noted registry
expectation.

Shortcuts that could be redirected were altered manually (not a
registry issue)

The only times I've altered the registry manually has been under the
direction of the 'owners' of the relevent line - such as in getting
Word2K to switch from unicode characters, or restoring stolen file
associations. Once I went on a rampage with Norton entries, to clean
out later revisions of their AV and other software.

RL

[legg]

On Sun, 17 May 2009 06:54:09 -0400, "dadiOH" <dadiOH@invalid.com>
wrote:

>legg wrote:
>
>> This left only the explorer issue, which persisted in the absence of
>> any system stress, but still triggered by user window manipulation.
>
>Excuse me butting in but have you tried using the Internet Explorer Repair
>Tool? I realize your problem isn't with IE but it and Explorer are quite
>intertwined and I've had success in flogging explorer into behaving by using
>the IE repair tool. It is available via MS System Information>Tools or
>Control Panel>Add-remove programs>Internet Explorer>Remove.
>
>dadiOH

I tried that finally this AM, after getting two Internet Explorer
error messages for the first time in some years, even though the
program wasn't running. It had been run to access a sticky website
earlier in the morning. These errors occurred consecutively with two
instances of the more familiar explorer.exe error, while transfering
file and opening windows. Other programs running at the same time,
were unaffected, as usual.

The repair option was unsuccessfull.
"Please run setup again to reinstall all component"
Details:
"Version 3.0.8168.0 of file atl.dll exists, but the version needs to
be greater than 3.0.8449.0"

3.0.8449.0 is available on the drive. The earlier one is ansi, the
later one unicode. Unicode........where have I heard that before.

RL

[legg]

On Sun, 17 May 2009 15:08:56 -0400, legg <legg@nospam.magma.ca> wrote:

>On Sun, 17 May 2009 06:54:09 -0400, "dadiOH" <dadiOH@invalid.com>
>wrote:
>
>>legg wrote:
>>
>>> This left only the explorer issue, which persisted in the absence of
>>> any system stress, but still triggered by user window manipulation.
>>
>>Excuse me butting in but have you tried using the Internet Explorer Repair
>>Tool? I realize your problem isn't with IE but it and Explorer are quite
>>intertwined and I've had success in flogging explorer into behaving by using
>>the IE repair tool. It is available via MS System Information>Tools or
>>Control Panel>Add-remove programs>Internet Explorer>Remove.
>>
>>dadiOH
>
>I tried that finally this AM, after getting two Internet Explorer
>error messages for the first time in some years, even though the
>program wasn't running. It had been run to access a sticky website
>earlier in the morning. These errors occurred consecutively with two
>instances of the more familiar explorer.exe error, while transfering
>file and opening windows. Other programs running at the same time,
>were unaffected, as usual.
>
>The repair option was unsuccessfull.
>"Please run setup again to reinstall all component"
>Details:
>"Version 3.0.8168.0 of file atl.dll exists, but the version needs to
>be greater than 3.0.8449.0"
>
>3.0.8449.0 is available on the drive. The earlier one is ansi, the
>later one unicode. Unicode........where have I heard that before.
>
I found the original installation setup files in the folder I keep all
communications software and programs - the prefered installation
directory when offered this option. I fear, therefore, that the repair
option may only effective for default-directory installations of this
program.

atl.dll was available in IE_S5.CAB. It is still ansi. With the old one
backed up and this single new file relocated into the system folder,
the repair option completed.

Explorer did it's little trick again, a few minutes after rebooting.

It will also close an explorer window when a file properties inquiry
is terminated, without generating a page fault message. This was
noticed while researching the properties of various atl.dll files
issued by other vendors, such as Adobe and Symantec, prior to the IE6
repair attempt.

RL

[dadioh]

legg wrote:
> On Sun, 17 May 2009 15:08:56 -0400, legg <legg@nospam.magma.ca> wrote:
>
>> On Sun, 17 May 2009 06:54:09 -0400, "dadiOH" <dadiOH@invalid.com>
>> wrote:
>>
>>> legg wrote:
>>>
>>>> This left only the explorer issue, which persisted in the absence
>>>> of any system stress, but still triggered by user window
>>>> manipulation.
>>>
>>> Excuse me butting in but have you tried using the Internet Explorer
>>> Repair Tool? I realize your problem isn't with IE but it and
>>> Explorer are quite intertwined and I've had success in flogging
>>> explorer into behaving by using the IE repair tool. It is
>>> available via MS System Information>Tools or Control
>>> Panel>Add-remove programs>Internet Explorer>Remove.
>>>
>>> dadiOH

> atl.dll was available in IE_S5.CAB. It is still ansi. With the old one
> backed up and this single new file relocated into the system folder,
> the repair option completed.
>
> Explorer did it's little trick again, a few minutes after rebooting.

I'm sorry it didn't help your problem.

dadiOH

[legg]

>On Sun, 17 May 2009 23:11:25 -0400, MEB <meb-not-here@hotmail.com> wrote:

>legg wrote:
>>> On Sun, 17 May 2009 03:59:10 -0400, MEB <meb-not-here@hotmail.com> wrote:
>>
>>>> legg wrote:
>>>> On Sat, 16 May 2009 22:40:07 -0400, MEB <meb-not-here@hotmail.com>
>>>> wrote:
>>>>
>>> large snip
>> <bigger snip>
>>> Well, we finally got to the registry modifications. Might have helped
>>> to advise of that from the start...
>>>
>>> So these were done manually?
>>>
>> Note that this was done over a year ago. It's just an example of the
>> methods used, and the records kept.
>>
>> Invalid keys were deleted through the Norton "WinDoctor" GUI. These
>> deletions are supposedly 'undo'able, though I've never reverted in
>> this manner - prefering to restore from registry back-up when needed
>> and to try again (or not).
>>
>> Missing dll and excutable files, if suitable ones be found, were
>> relocated (copied) manually to correspond to the noted registry
>> expectation.
>>
>> Shortcuts that could be redirected were altered manually (not a
>> registry issue)
>>
>> The only times I've altered the registry manually has been under the
>> direction of the 'owners' of the relevent line - such as in getting
>> Word2K to switch from unicode characters, or restoring stolen file
>> associations. Once I went on a rampage with Norton entries, to clean
>> out later revisions of their AV and other software.
>>
>> RL
>
>Okay, glad you had posted that you HAD made registry modifications and
>you had kept track, and that these are old, hence potentially not
>related to the present condition.
>
> Not knowing where you are on your backup procedures:
> *IF* you have a recent version from prior to the errors, you might try
>exporting the backed-up registry files for a trial at replacement. OR,
> *IF* you have a back-up from before [prior to] the issue and you
>haven't added anything recently [and no major hardware changes], you
>might be able to try just that much of a "repair" [the user.dat and
>system.dat] or pull one of the ones from the automatic back-ups done
>from %windir%/sysbckup/rb*.cab [not the bad if there is one]. Modify the
>date to one within your present backup dates and within the 1-5 format
>available.
> Save the present before replacement and compare the differences
>between the two sets.
>
> *OR* you're back to "trial and error", or using the diagnostics I
>previously provided.
>
> The issue of the one hour file difference is something you will need
>to research as you have negated our attempted explanations and directions.
>
> *IF* you wish, you might post the file names [system only] that you
>think are at issue, along with OS installation date and other that might
>be relevant. You checked for similar dates [changed date] within the
>system, right?
>

I ignored last-mod timestamps with the single hour difference in SFC.
I don't know if this will prevent their future identification, using
the same tool. Haven't done anything except make a few notes and run a
few hard drive tests, lately.

In doing so, there was an explorer.exe page fault (Kernel32) in safe
mode - a first. This was in the process of closing a running program -
Conversions Plus Quickview - looking at the ansi text file contents of
'default.sf0'. As in normal mode, all this fault did was close windows
and loose window viewing preferences, effecively restarting
explorer.exe.

This might reduce candidate sources for the error, at least to those
supported by or actively supporting safe mode. These parts of the
operating system might be expected not to change much, possibly
increasing the usefulness of a past back-up. It also indicates that
file transfers will have to be made from outside of the environment ~
easiest option being to use the secondary boot W2K OS.

Even if these errors eventually are avoided, there will likely be new
issues surrounding the use of the IE repair tool. The final act of
this 'repair' redeposited an IE5.5SP2 link icon on the desktop. This
is an icon I recall naming many years ago, on the day of installation
of that version. IE6 and the last KBs/security patches were installed
over 3 years ago. I hate to think what else might have been undone.

IE6.1 at the pre-repair rev still runs from the toolbar. Perhaps this
is one of the benefits (?) of using the non-default installation
directory. The original installation and update cabs for IE6 and its
KBs remain in this directory.

I will ignore Internet Explorer issues until explorer.exe behavior is
dealt with.

RL

[legg]

>On Mon, 18 May 2009 14:09:56 -0400, legg <legg@nospam.magma.ca> wrote:

>>On Sun, 17 May 2009 23:11:25 -0400, MEB <meb-not-here@hotmail.com> wrote:
>
>>legg wrote:
>>>> On Sun, 17 May 2009 03:59:10 -0400, MEB <meb-not-here@hotmail.com> wrote:
>>>
>>>>> legg wrote:
>>>>> On Sat, 16 May 2009 22:40:07 -0400, MEB <meb-not-here@hotmail.com>
>>>>> wrote:
>>>>>
>>>> large snip
>>> <bigger snip>
>>>> Well, we finally got to the registry modifications. Might have helped
>>>> to advise of that from the start...
>>>>
>>>> So these were done manually?
>>>>
>>> Note that this was done over a year ago. It's just an example of the
>>> methods used, and the records kept.
>>>
>>> Invalid keys were deleted through the Norton "WinDoctor" GUI. These
>>> deletions are supposedly 'undo'able, though I've never reverted in
>>> this manner - prefering to restore from registry back-up when needed
>>> and to try again (or not).
>>>
>>> Missing dll and excutable files, if suitable ones be found, were
>>> relocated (copied) manually to correspond to the noted registry
>>> expectation.
>>>
>>> Shortcuts that could be redirected were altered manually (not a
>>> registry issue)
>>>
>>> The only times I've altered the registry manually has been under the
>>> direction of the 'owners' of the relevent line - such as in getting
>>> Word2K to switch from unicode characters, or restoring stolen file
>>> associations. Once I went on a rampage with Norton entries, to clean
>>> out later revisions of their AV and other software.
>>>
>>> RL
>>
>>Okay, glad you had posted that you HAD made registry modifications and
>>you had kept track, and that these are old, hence potentially not
>>related to the present condition.
>>
>> Not knowing where you are on your backup procedures:
>> *IF* you have a recent version from prior to the errors, you might try
>>exporting the backed-up registry files for a trial at replacement. OR,
>> *IF* you have a back-up from before [prior to] the issue and you
>>haven't added anything recently [and no major hardware changes], you
>>might be able to try just that much of a "repair" [the user.dat and
>>system.dat] or pull one of the ones from the automatic back-ups done
>>from %windir%/sysbckup/rb*.cab [not the bad if there is one]. Modify the
>>date to one within your present backup dates and within the 1-5 format
>>available.
<SNIP>

Good news.

After importing a manual registry backup from January ( last one
guaranteed to be from before the first crash in April), the
explorer.exe page faults appear to have ceased. I'll fool around for
another week before going through with a belated system back-up. The
resulting registry is still larger than it's predecessor or the
original Jan version, so there's still stuff remaining from before the
over-write - but it appears to be not whatever was screwing things
around in simple window management.

There are some benefits, it seems, to being a SW reactionary - having
made so few changes to the machine or software in the intervening
period. The microsoft registry restore dates, accumulated over a very
short period, recently, could not do this. I expect they'd also be
useless, in this situation, to anyone who simply turned their machine
off daily.

Anyways, this is much less labour intensive than building up a
previous HDD with intervening work files, so kudos is deserved all
around.

I'd like to thank one and all for their patience. You never learn
anything from a re-install.

Anyone care to speculate on what registry lines could give
explorer.exe the symptoms of virtual memory errors, like the ones
experienced here?

RL

[buffalo]

legg wrote:
> Anyone care to speculate on what registry lines could give
> explorer.exe the symptoms of virtual memory errors, like the ones
> experienced here?
>
> RL
I'm sure there are some free programs that can compare two Registrys and
point out the differences.
That should cut it down to a reasonable level.
Buffalo

[buffalo]

MEB wrote:
[snip]
> YEEeeeeHAAAaaaa,,, glad to hear it.
>
> Really, there are so many potentials with these types of errors
> involved, either diagnostics or the suggested cross-comparison check
> are the only ways to determine your causal issue *IF* you've already
> done the preliminary checks [as you had].
>
> BE CAREFUL "foolin around", you got lucky this time. At least you
> now have some recovery/diagnostic techniques to analyse and use next
> time.
>
> Just curious,,, what's running QFE show now?

Just gotta love Registry Cleaners because they really speed things up!!!

[pcr]

MEB wrote:
> legg wrote:
>> On Sat, 16 May 2009 17:07:30 -0400, MEB <meb-not-here@hotmail.com>
>> wrote:
>>
>>> legg wrote:
>>>> On Fri, 15 May 2009 22:11:41 -0400, MEB <meb-not-here@hotmail.com>
>>>> wrote:
>>>>

....snip...
>>>>> 2. Run QFE to see if your updates are still found/shown as
>>>>> installed. Note potential issues.
>>>> Running QFE on this machine simply opens the QFE folder window,
>>>> with the W98 folder.
>>> HMM, were all your updates from local network sources? Ah, wait,
>>> was there a plus sign next to the folder? If it didn't open AND you
>>> have sub folders then you likely have no errors. It also contains
>>> the ability to actually check for installed files verses the QFE
>>> updates [second tab].
>>
>> No plus sign. Any updates were MS Update, on-line.
>
> You may have lost those entries from the registry.
> Have you run any registry cleaning tools on the system or had
> registry errors in which you attempted fixes?
> IF SO, what types/fixes?
>


I think what was intended was QFECheck...

QFECheck.exe
Desc: Windows 98 Update Information Tool
Loc: C:\WINDOWS
Size: 36,864 bytes
Mod: Monday, July 27, 1998 03:48:26 PM
Ver: 4.10.1998

....snip...
--
Thanks or Good Luck,
There may be humor in this post, and,
Naturally, you will not sue,
Should things get worse after this,
PCR
pcrrcp@netzero.net