Re: insufficient key column information

Billy Verreynne <vslabs_at_onwe.co.za> wrote in message news:<b3ncfd$kmm$1_at_ctb-nnrp2.saix.net>...
> Scott wrote:

Thanks for responding Bill,

> Scott, that's what happen when you take a stockcar rider (VB code running on
> MS Access) and put him on a state-of-the-art F1GP car. :-)
>

I'm not sure that that's the type of problem I'm having. In theory, I should be able to plug any supported db into an application using ADO, and it should work. Although I expected to have to tweak things, I never expected the problems I've had. For example, duplicates are easily handled by Access and SQL Server 7 because the recordset brings table relationships into the recordset. So using tableA.id and tableB.id will result in different members of the recordset being accessed. However, in Oracle, this doesn't happen. It will not make the distintion and will only find the first id. This is why I had to alias all the duplicate column names.

> Back in '94 I did my own port of a VB app I did, from MS Access, to
> SQL-Server 4. Performance sucked (and a single app instance loaded,
> generated 50% of the total LAN traffic).
>

Did you ever discover where the loss in performance came from?

> I winded up re-writing the _entire_ application, basically from scratch.
> Through the years I've seen others running into this same problem time and
> time again. So be very wary when it comes to functionality and performance
> of your VB/MS Access app running on Oracle.
>

I'm about to start from scratch (I would have it done by now)

Thanks

Scott Received on Fri Feb 28 2003 - 12:45:05 CST