Access 2003 Crash

[JK Peck]

I have a fairly large Access application that ran correctly in Access
2000. After upgrading to Access 2003 (and recompiling and updating
references), it reliably crashes at a certain point. If I step
through the VBA code, the crash does not occur.

What is different about stepping through code instead of just running
it?

Any idea how to find the cause? I know about where it happens, but
since it is Access itself crashing, finding a workaround is just trial
and error.

[Larry Linson]

Have you used the Office Update to assure that you have all the updates to
Office 2003 (they'd be patches, rather than Service Packs, at this point)
and to Jet 4.0 (it's up to Service Pack 8, for Office 2003)?

"Access itself crashing" isn't normal behavior. I can assure you that is not
in the category of "working as designed". Since there _should_ be nothing
the user can do to cause this, I'd suspect it would be a free incident with
the support team (you could call and ask).

Unless they have a patch that hasn't yet been fully tested and released that
applies, however, I am reasonably sure that one of the things they will
suggest is that you first uninstall Office/Access, and then re-install it,
then run Office update. Whether that will fix the problem or not, it's
something that usually has to be done to eliminate an install error. And,
it's entirely possible that whatever is triggering the crash is some
external function that Access is using, so it might not be something in
Access or Office.

Timing is different when you step through code and when you run it. From a
user point of view, though, there's not really anything we can do about
"timing problems" inside the host (Access, in this case).

But, if you can narrow down where it is happening, if that is in code, and
you post some code around that area, with your best guess at where it
happens, maybe someone could suggest another way to do it that _might_ not
trigger the same consequences. Don't count on that, but it's a possibility.

Larry Linson
Microsoft Access MVP
"JK Peck" <pe**@spss.co m> wrote in message
news:bc******** *************** ***@posting.goo gle.com...

I have a fairly large Access application that ran correctly in Access
2000. After upgrading to Access 2003 (and recompiling and updating
references), it reliably crashes at a certain point. If I step
through the VBA code, the crash does not occur.

What is different about stepping through code instead of just running
it?

Any idea how to find the cause? I know about where it happens, but
since it is Access itself crashing, finding a workaround is just trial
and error.

[J Peck]

I did check for updates, and I found none available. My Access version
is 11.5614.5703, and Jet 4.0 SP8.

I suspected a race condition or other timing problem and placed some
alerts in the suspect areas to make the application pause several times
in the code, but the crash still occurs. Some random permuting of code
gave the same result, unfortunately.

I wondered whether single stepping bypassed any optimization in the code
(if there is any when compiling). But I can't really ask users to
single step the code in order to get their job done!

This has occurred on two different recent Access 2003 systems, one on
Win 2K and one on WinXP.

*** Sent via Developersdex http://www.developersdex.com ***
Don't just participate in USENET...get rewarded for it!

[Tony Toews]

pe**@spss.com (JK Peck) wrote:

I have a fairly large Access application that ran correctly in Access
2000. After upgrading to Access 2003 (and recompiling and updating
references), it reliably crashes at a certain point. If I step
through the VBA code, the crash does not occur.

Try inserting some DoEvents in the code.

Tony
--
Tony Toews, Microsoft Access MVP
Please respond only in the newsgroups so that others can
read the entire thread of messages.
Microsoft Access Links, Hints, Tips & Accounting Systems at
http://www.granite.ab.ca/accsmstr.htm

[Larry Linson]

"J Peck" <pe**@spss.co m> wrote

The "build information" you cite is identical to mine, and I haven't had any
similar problems.

This has occurred on two different recent
Access 2003 systems, one on Win 2K and
one on WinXP.

Was this a copy of the same database? Is it happening in an MDB or in an
MDE? Have you tried adding a few DoEvents in the area of code that is
suspect?

Perhaps it is some form of corruption. To get around _that_, the best way
would be to compact and repair your database. If that doesn't help, create a
new database, copy in all the objects _except_ the ones you suspect as
causes, and re-create those suspect ones in the new database.

If that doesn't do it, and someone else doesn't have some suggestions, I
think you may need to contact M'soft support. As I said, with Access itself
crashing, that is not the way it should operate and the user should not be
able to cause it from the database the user creates.

Larry Linson
Microsoft Access MVP

[David W. Fenton]

pe**@spss.com (JK Peck) wrote in
news:bc******** *************** ***@posting.goo gle.com:

I have a fairly large Access application that ran correctly in
Access 2000. After upgrading to Access 2003 (and recompiling and
updating references), it reliably crashes at a certain point. If
I step through the VBA code, the crash does not occur.

Have you decompiled and recompiled?

Have you created a fresh MDB and imported everything into it (from a
decompiled version)?

I've seen problems with compiled code causing crashes with slightly
different versions of the VBA libraries, and decompiling always gets
rid of it.

Of course, in A2K, decompiling is not quite as easy as in A97 (it
doesn't clean out as much gunk), hence the suggestion about
importing into a new MDB.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

[Larry]

pe**@spss.com (JK Peck) wrote in message news:<bc******* *************** ****@posting.go ogle.com>...

I have a fairly large Access application that ran correctly in Access
2000. After upgrading to Access 2003 (and recompiling and updating
references), it reliably crashes at a certain point. If I step
through the VBA code, the crash does not occur.

What is different about stepping through code instead of just running
it?

Any idea how to find the cause? I know about where it happens, but
since it is Access itself crashing, finding a workaround is just trial
and error.

When does this happen? Sound similar to what was happening (Access
2000 and XP both) in my code when opening a form with lots and lots of
subforms. The way I found around it was to leave the subforms with no
object. Then I put a .5 second delay to run the form timer event and
set up the source object there. If you do this and it solves your
problem, you'll have to pay attention to the child and master links.
Access seems to trash them if you have more than one field in the
link.

I hope this helps.

[JK Peck]

Thanks for all these suggestions. After trying all the usual,
including DoEvents to no avail, I inserted SysCmd messages to the
status bar and found that the line of code that triggered the crash
was simply
Me.refresh
after the code had made various updates. And, indeed, commenting out
that line and running Refresh from the menu once the code had
completed did crash.

Now the form in question consists of a main form and a subform. The
code in question only affected the subform, so I created a refresh
function in the subform and called that from the main form code, and I
get the refresh without a crash. The data tables, which are actually
in a linked database, were already getting properly updated.

This suggests to me that the form data structure in memory has gotten
corrupted somehow, but this has at least moved the problem far enough
away that we can get our work done.

[David W. Fenton]

pe**@spss.com (JK Peck) wrote in
news:bc******** *************** ***@posting.goo gle.com:

This suggests to me that the form data structure in memory has
gotten corrupted somehow, but this has at least moved the problem
far enough away that we can get our work done.

My suggestions would get rid of any form corruption.

And it's why I posted them, because form corruption can lead to
Access crashes.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

[JK Peck]

"David W. Fenton" <dX********@bwa y.net.invalid> wrote in message news:<Xn******* *************** ************@24 .168.128.78>...

pe**@spss.com (JK Peck) wrote in
news:bc******** *************** ***@posting.goo gle.com:

This suggests to me that the form data structure in memory has
gotten corrupted somehow, but this has at least moved the problem
far enough away that we can get our work done.

My suggestions would get rid of any form corruption.

And it's why I posted them, because form corruption can lead to
Access crashes.

I would have gotten rid of the corruption if I could, but compact and
repair and even exporting everything to a clean database does not
solve it. I think it is happening dynamically.

While the code change I made gets around the problem, the corruption
is demonstrably still there, since doing a Refresh from the menu bar
still causes the crash.

In fact, I found that as soon as the form is opened, it is corrupted,
i.e., Refresh crashes it. There is a little bit of onLoad and
onCurrent code that is run, but it is pretty innocuous. I'll be
trying to pin this down further.