From: planetthoughtful on
Hi All,

I have an application that I need to make as 'user proof' as possible.

Current roadblock is that one of the operations it performs does a
series of updates, inserts etc on tables on our SQL Server, and this
occasionally fails as a result of one or more records being deleted
from the recordset during normal user interaction via other tools.

I'm wondering how to best recover from this (currently, because I
understand the underlying code, I simply drop into debug, wait a few
seconds, press F5 to keep running)?

I assume it's best practise to use some sort of ON ERROR condition,
but I could see fringe scenarios where it might get into a very long
loop if it keeps hitting the same condition every time it loops back
to re-attempt the operations.

Any suggestions very welcome.

pt
From: Roger on
On Jul 25, 8:10 pm, planetthoughtful <planetthought...(a)gmail.com>
wrote:
> Hi All,
>
> I have an application that I need to make as 'user proof' as possible.
>
> Current roadblock is that one of the operations it performs does a
> series of updates, inserts etc on tables on our SQL Server, and this
> occasionally fails as a result of one or more records being deleted
> from the recordset during normal user interaction via other tools.
>
> I'm wondering how to best recover from this (currently, because I
> understand the underlying code, I simply drop into debug, wait a few
> seconds, press F5 to keep running)?
>
> I assume it's best practise to use some sort of ON ERROR condition,
> but I could see fringe scenarios where it might get into a very long
> loop if it keeps hitting the same condition every time it loops back
> to re-attempt the operations.
>
> Any suggestions very welcome.
>
> pt

what happens if you put all your insert/update/delete statements
within a 'transaction' ?