Same form, but one system GPFs...?

On Wed, 20 Feb 2008 19:35:15 +0100, "Anders Jonsson"
wrote:

>
>> 100% it is (and will be from now on) but it is maddening
>> that today I could not generate the GPFs that were plaguing
>> me yesterday. Yesterday, it was intermittent in the sense
>> that it would GPF every two or three times I would click on
>> the button. Today, I must have run the button fifty times
>> with no problem. Nothing I can detect had changed...
>>
>
>
>I don't understand what youare saying. Before you had "Best fit" - now you
>have "100%", how can you then say that nothing has changed?
>
>As others have said - running at anyting but 100% could cause strange
>problems. It is very likely that you finally have found the cause!
>
>Anders
>
>
Hi Anders,

Perhaps I described the order of events poorly...

I ran the code on form today many, many times with no GPFs
(even though yesterday I got GPFs nearly every time I ran
the code on the form.)

That happened even though "nothing had changed" on the
system, or the form.

After that, I changed from Best Fit to 100%.

Of course it may turn out that the Best Fit was the cause of
the problem, but the problem was intermittent, and so I have
no way to know that at this point because just prior to
changing to 100% I could not make the GPF happen.

I hope that clarifies, and thank you and others for all the
help,
--
Kenneth

If you email... Please remove the "SPAMLESS."
 >> Stay informed about: Same form, but one system GPFs...? 

> As others have said - running at anyting but 100% could cause
> strange
> problems. It is very likely that you finally have found the cause!

Doesn't really matter.

Using the 'c' functions in a network environment is known to
*potentially* cause issues.

Using other than a '100%' ZOOM factor is known to *potentially* cause
issues.

If the whole exercise only got those two things resolved, I believe it
will prove to have been worth the time/thought/digging it has taken to
gotten there. That the issue appears to have been resolved is a
bonus.

And it doesn't even matter that only one of several systems was
exhibiting the behaviour; too many things could come into play.
Accept that it is fixed on that system (if it is), and that the others
didn't need to be 'fixed'.

--
------------------------------
Tony McGuire
 >> Stay informed about: Same form, but one system GPFs...? 

> I hope that clarifies,

Yes it does.

Anders
 >> Stay informed about: Same form, but one system GPFs...? 

> Using the 'c' functions in a network environment is known to *potentially*
> cause issues.

The "only" problem with cMax that I know is that it requires a full lock
which of course could be problematic on a shared table. However, I have
never heard that it has cause a GPF and using it against a table in PRIV
that is not being edited shouldn't be a problem. At least that is how I have
understod the issue.

>
> Using other than a '100%' ZOOM factor is known to *potentially* cause
> issues.

Agree.


> Accept that it is fixed on that system (if it is), and that the others
> didn't need to be 'fixed'.

Well as this is a discussion forum and the traffic isn't exactly heavy I
find it interesting to discuss a problem and find a plausible solution. IMHO
the zoom factor or the path to PRIV is much more likely cause of GPF than
the use of cMax.

But you are right, if it's working now, why bother.

Anders
 >> Stay informed about: Same form, but one system GPFs...? 

> The "only" problem with cMax that I know is that it requires a full
> lock
> which of course could be problematic on a shared table. However, I
> have never heard that it has cause a GPF and using it against a
> table in PRIV that is not being edited shouldn't be a problem. At
> least that is how I have

Except that when the issue was presented he wasn't using :priv: - he
changed it later.


> Well as this is a discussion forum and the traffic isn't exactly
> heavy I find it interesting to discuss a problem and find a
> plausible solution. IMHO

Well, I think we did that. I was hoping only to avoid the 'why does
Paradox act that way', which I figured was next, when there is no one
here who could even begin to answer.



--
------------------------------
Tony McGuire
 >> Stay informed about: Same form, but one system GPFs...? 

Anders Jonsson wrote:
>
> But you are right, if it's working now, why bother.

Because it could, potentially, stop working again?

I prefer to find the cause of a problem rather than ignore it just
because it's gone away. Just because you can't see the Crocodile doesn't
mean it's not there, lurking ... waiting in the reeds for a naive
tourist to go for a swim ...

Unfortunately, I have no more ideas on this issue other than to
uninstall and re-install Paradox and the BDE, using a good, commercial
registry cleaner in between.

TomK
 >> Stay informed about: Same form, but one system GPFs...? 

On Thu, 21 Feb 2008 19:42:32 +1100, Tom Krieg
wrote:

>Unfortunately, I have no more ideas on this issue other than to
>uninstall and re-install Paradox and the BDE, using a good, commercial
>registry cleaner in between.
>
>TomK

Hi Tom,

About three months ago, I had to re-install Windows on the
machine in question.

When the installation was complete, I made an image of the
system.

Then, as I installed the system's software, I repeated the
image creation process a total of six times.

When this GPF problem started, as part of my diagnostic
effort, I restored the system to the earliest image that had
Paradox on it, in effect, doing what you suggested above.

Your idea is perfectly reasonable, but unfortunately, did
not solve the problem.

At this time, because the problem seems truly intermittent,
I suspect that it is hardware, rather than software related.
But (though this may be hard to imagine) I understand even
less about that <g>.

I have been experimenting with the form every which way and
have not been able to generate the GPF in the last day.

My Assistant will actually be using it tomorrow, so we shall
see...

Sincere thanks to all for the help,
--
Kenneth

If you email... Please remove the "SPAMLESS."
 >> Stay informed about: Same form, but one system GPFs...? 

On Thu, 21 Feb 2008 07:23:09 -0700, "Tony McGuire"
wrote:

>> The "only" problem with cMax that I know is that it requires a full
>> lock
>> which of course could be problematic on a shared table. However, I
>> have never heard that it has cause a GPF and using it against a
>> table in PRIV that is not being edited shouldn't be a problem. At
>> least that is how I have
>
>Except that when the issue was presented he wasn't using :priv: - he
>changed it later.

Hi Tony,

(and no, I am not moving in the direction of "Why does
Paradox work that way...<g>)

Just to clarify:

I had not modified anything relating to the :PRIV: in my
code.

I was trying to make any modifications is a systematic
manner, and the change of the :PRIV: was next on the list,
but I had not made that change when the GPFs stopped.

As was true when I had the frequent GPFs, right now, the
form is on the server, it has a button that runs a query,
the answer.db is local to the machine with the problem, and
that machine's :PRIV: is the default (and has 54 characters
in the path string.)

All the best,
--
Kenneth

If you email... Please remove the "SPAMLESS."
 >> Stay informed about: Same form, but one system GPFs...?