--- In IBObjects@yahoogroups.com, Aage Johansen <aagjohan@o...> wrote:
<snip>
------------------------------
>
> Is there anything in particular that we (as IBO developers/users)
> should look out for?
As one of the test case providers of the bug in question, I have to
say that there are a number of symptoms to this particular bug which
has made it hard to track down, as I understand it. I spent about
10 hours trying to eventually confirm that it ended in an API call
in the client dll, everything upto that point always looked good as
far as I could tell.
In my case the bug manifested itself only after we had started using
the 1.5 release, but I've since found out that it can be found in the
1.0x series and even into Interbase. The problem appears, at least
for me, to be releated to TCP/IP connections, since if we went local
the application in question had no problem. It also only appeared in
a very specific set of steps. Once the bug triggered (as an exception
in GDS32.DLL) then the entire application would then freeze
solid. I would also like to mention that while we were on the 1.0x
release we never saw this behaviour. Although if you talk to Rohit
Gupta, his group has been trying to figure it out for a number of
months, both on FB (1.0x & 1.5) as well as on Interbase (< 7.x) I
believe.
The (short-term?) solution that Nickolay presented us was to run with
the IB_Connection.CacheStatementHandles set to false. This appears
to fix the bug for the time being but it's hard to tell yet what kind
of performance impact that may or may not have on our connections.
I personally would like to see more work/information done to see this
bug fixed, either in the client dll or in the IBO code.
At this time I would just like to say that I appreciate all the work
that Nickolay did with tracking this down so quickly.