>Hi Peter,
>
>we had a similar problem a few years ago with a standalone FP 2.6 application under Win98 (no Novell). The environment is quite different to the one you discribed, but the corrupted tables looked exactly like yours.
>
>Usually the corruption occured in the morning, while a stack of bills where calculated and printed, but I don't recall whether there always had to be midnight between the last successful writing process and the one that raised the problem. We never found the cause of that problem, though it disappeared after implementing a periodical (once a week) deletion und rebuilding of the index files.
>
>Robert

Robert, this looks interesting anyway, because with my so-many tests to stress the multi-user environment (over a midnight) we were never able to recall the thing at will. Interesting as well, because you made me think of a time-problem in Fox around midnight (clock encounters a time warp).
I am not sure how this can be related, but it's in my mind now.
I think too about the fact that once the server has been re-started, it about always stays away for some weeks ...