There are a number of ways to implement pessimistic or optimistic buffering in SQL Server.

1.) ADO supports both optimistic and pessimistic buffering. RTM on MSDN for this info. It's easiliy done. Create a middle tier in ADO and have VFP talk to that.

2.) Roll your own by using semaphore locks. Create a column in every table or a separate locking coloumn that holds a flag to "locked" tables and or rows.

Both have pros and cons. There is no one right answer. FWIW on our current project we use optimistic buffering with timestamps. We do not require the granularity of pessimistic locks - but you might. Each app is different.

Be careful what you wish for - you just might get it. Locking in SQL is more involved than in the fox world. Make sure you understand the concepts of isolation levels and transactions before going to "far out of the norm"

Hope this helps, Val Matison


>Hi guys!
>
>Does any body know where may I find an article on row level locking under Client/Server Enviorments?
>
>Maybe with a sample of a case where user A cannot access record Number 31 because user B is currently using it with read/write intentions.
>
>Thanks in advance.
>
>NJM