Mika Raento wrote:
> I somehow read that to indicate that you need the management node after
> the connection. What it tries to explain is that the cluster-connection
> is bound to a _cluster_ not a management node. Maybe a documentation bug
> should be filed?
I've sent an email to one of the doc guys.
>> I do not yet have the answer... but do you have and code that
>> illustrates a problem with how object lifecycles are being done
>> incorrectly? I'd love a test case so we know what we are working with.
>
> Well, something as simple as letting the transaction go out of scope
> when posting a executeAsynchPrepare will cause a segmentation fault.
> Change the async.pl code so that it doesn't reference the transaction
> object in the callback sub and you'll see it.
Good point. I'll use that as a test case and see what we can do.
By the way -the %ref and %unref stuff I emailed in earlier doesn't seem
to solve our problems after all...
--
Monty Taylor
Senior Consultant
MySQL Inc., www.mysql.com
Get More with MySQL! www.mysql.com/consulting
