> Is there some chance you could create a reproducible test case?

Hi Milan,

Sending you a copy of the database would not be a problem. Obviously
what you would then want is a set of SQL statements to run to emulate
the front-end software, so I just tried to do that, and everything
worked as expected (ie no problems).

Have you by any chance seen this problem before when the front-end was
using DAO/ODBC to set the new field values? The front-end in this case
sets the new field values one-by-one using traditional DAO techniques
(yes, I know, that's just the way it evolved over the years). If I
replace the field update with a query:

UPDATE "Readers" SET "AlwaysMonitor" = 0;

the problem goes away. So thankfully it looks like I've found a good
solution, but it's curious that setting the values one-by-one through
DAO did set the values, but apparently corrupted the Firebird database
in some way (MySQL isn't similarly effected). Would it be helpful if I
sent you two databases, one before the front-end has executed the
changes, and one after it has executed them, to allow you to see what is
the nature of the corruption?

Regards,

Robert.

>
>
>
> ------------------------------------------------------------------------
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.5.339 / Virus Database: 270.12.34/2122 - Release Date: 05/19/09 06:21:00
>

----------


----------


No virus found in this outgoing message.
Checked by AVG - www.avg.com
Version: 8.5.339 / Virus Database: 270.12.34/2122 - Release Date: 05/19/09 06:21:00


[Non-text portions of this message have been removed]