>Yes, that'll prevent the hardware from returning a "successful" prematurely to the OS.
>
>But how to get the OS to NOT return immediately to VFP (the OS queueing it's own FLUSHes for later execution), that is the question!

Actually, this be done in VFP code. See Nathaniel Moody's post at Message #647125 for sample code using the FlushFileBuffers winapi function.

However, I don't know what the performance impact on the server will be if multiple clients are using it regularly. This is part of the research we need to do before we can add the FlushFileBuffers call to a future version of VFP.

Thanks for adding to the Wishlist!

-Tom

[This post is provided AS-IS with no warranties, and confers no rights.]