Warren Young wrote:
>> I'm wondering if anyone has gotten this message when running the
>> mysql++ samples. It's occuring when main() exits and the system is
>> trying to release Connection con.
>Try simplifying the code to the point that the error goes away. This
>may prove insightful.
I've isolated it to the release of the Connection object. For example,
if I change it to Connection*, new it,and DON'T delete it (thus leaking
it), I experience no exception/crash.
>> My problems seem to be machine-specific,
>Indeed, it's compiler-specific as well. Keep in mind, MySQL++ is
>cross-platform.
I guess I should have been more specific--it is machine-specific in that
I have the same platform and setup (WinXP, etc.) working successfully
on several other machines--it's just this one particular computer that
is giving me trouble.
>> Specifically, I'm running myql 5.0.0-alpha
>I can't have too much sympathy with your problems since you're running a
>pre-stable version of the database server, especially since you're able
>to crash it. Is there a good reason why you aren't using 4.0 or 4.1?
Your lack of sympathy notwithstanding, I should have mentioned that I had
tried reverting to the production version unsuccessfully (hence retrying
the example programs) on this one particular computer. My reasons for
testing against 5.x? Wanting to leverage the new functionality.
>> combinations of mysql and mysql++. For example, if I'm running mysql
>> 5.0.2-alpha and mysql++ 1.7.25, the MySQL service terminates.
>I can't imagine a MySQL++ error that could cause MySQL to die. A
>non-buggy version of MySQL should be uncrashable with any conceivable input.
I guess I'm looking for someone who has had the same specific problem that I
have (again, it seems to be a computer-specific WinXP issue, so is it some
Windows registry thing? is it a Win32 application conflict thing?), not
someone who cannot imagine/conceive of the problem--I do, however, appreciate
your first suggestion.
