Hi,

> It is very possibly related with isc_tpb_consistency mode if it is
> used.

Why should it? If this statement is executed using one connection
there is no concurrency here. Statements are executed one by one,
within a context of one transaction in non-autocommit mode, or
different but serially ordered transactions.

If statements are executed each in its own connection (like get
connection from the pool, execute statement, return connection to the
pool), such problem cleanly points out on the bug in either pool
implementation, or our transaction management. That is why I asked for
the code that causes this bug.

Also, I do remember that engine had some problems with "object is in
use", but I do not know if they were corrected in FB 1.0 or not. As a
workaround people suggested to execute DDL statements in autocommit mode.

Best regards,
Roman Rokytskyy