Hi!

> So, I've done some testing (using FlameRobin as client) and found these:
>
>
> 1) When a query fail, ex:
>
> select cast('A' as integer) from rdb$database
>
> it does NOT generate 2 EXECUTE_STATEMENT_FINISH (a 'normal' followed by a FAILED). It just generate a FAILED one (as expected).
>
>
> 2) I tested the one very simple statement that happens to *always* generate the 2 EXECUTE_STATEMENT_FINISH (normal folowed by FAILED) with FlameRobim:
>
> SELECT * FROM CONFIGURACAO
>
> AND there is no error! no exception! everything seems just normal, but... the second FAILED EXECUTE_STATEMENT_FINISH is always there!

Does this also happen with other tables or just the one above?

> 3) During this and some other tests, I found out that the fb audit IS NOT writing every statements to the log file. Sometimes you execute statements that NEVER shows on audit log file -- it happened executing from the application, and from FlameRobin, on different ocasions! Sort of frustrating!
>
>
> Any thoughts on that?

It would be interesting if you have been running a user trace or a
system audit and with what trace configuration.

In respect what a user trace sees, the filter properties and under what
Firebird user the trace session has been started is important.


--
With regards,

Thomas Steinmaurer
Upscene Productions
http://www.upscene.com
http://blog.upscene.com/thomas/

Download LogManager Series, FB TraceManager today!
Continuous Database Monitoring Solutions supporting
Firebird, InterBase, Advantage Database, MS SQL Server
and NexusDB!