Hi Len,

What are the chances that you are running out of RAM (real or virtual)? Who knows how such a situation might exhibit itself, especially if you are not completely out of the stuff?!?!?!?

At a minimum it would look like you are creating at least 210 objects in your 35 minute test, and if these are hanging around rather than being released then RAM could be the issue.

Keep in mind that a RELEASE will do NOTHING of significance, even though it will appear to run correctly if you have a "dangling reference" hanging about. The memvar with the object reference will be removed but the object itself will remain.

Good luck,

Jim N

>I ran this once and got 0 replies, so here it is again. Has anyone ever encountered a random syntax error during a CREATEOBJECT() call. The error occurred only once during 35 minutes of testing that took only 10sec per test. The error occurs on the development machines as well as 20 of 70 production machines, networked as well as standalone, and appeared only after we converted from vfp3 to vfp6 (the SP3 did not remove it). Even a method in which to debug the problem would be useful. In fact, running the EXACT command from the command line after a SUSPEND on the error, runs correctly. It has been extremely ifficult to duplicate the problem on demand. The parameters being passed into the function check out fine. Any and all help will be appreciated.
>
> -- Len