Hi Helen,

> Unfortunately, Roman, this doesn't help. Firebird 1.5 doesn't have
> any gdscodes in the 3365... range so, even if you're somehow able to
> generate this range from msgs.fdb in the Fb 2 tree, it doesn't give
> much of a clue about which exception is actually occurring in Fb 1.5.

I get from the engine an error code, which is then looked up in the message
file (much more simpler structure than firebird.msg), so somehow server was
able to generate the error.

> From ignorance of how the Jaybird driver versions map to the
> Firebird server versions, I wonder whether Chris' problem is somehow
> related to a mismatch between the Jaybird client and the Firebird
> server...?

Very unlikely. Jaybird 1.0.0 is pure Java version implementing the v10 wire
protocol (the one introduced in IB 6.0) and does nothing specific there. It
is proven to run with IB 6.0, FB 0.9.x, FB 1.0.x and turns out - with FB
1.5.x. It does not load any Firebird code, just sends and receives data over
the TCP socket.

But anyway, JDBC driver just received a raw status vector (byte array) and
extracted the error code from it. Then it just looked for an appropriate
message in error file. But even if error file was not correct, the error
code is the one returned from the server.

Now, if you say that FB 1.5 could not have 3365.... error codes, we have to
check how did it happen. I will get the FB 1.5 code base and appropriate
msg.fdb and check what is there.

Roman