Subject Re: Exact timestamp w/o using UDF?
Author Adam
--- In, "Joe Martinez" <joe@...> wrote:
> Thanks for all the quick replies.
> At this point, I think I'm just going to go ahead and use the
> getExactTimestamp UDF. It's not a difficult to declare as I'd
> remembered, and it will be good enough for what I'm doing. I don't
> necessarily need it to be unique for the entire database, as long as
> sequential operations from one workstation are each unique. For
> example, if a client does an update, and then a delete, I just need to
> know that the update happened before the delete, and ten thousandths
> of a second should be small enough for that.

Well if you are not going to fix the design, do a sleep for 1ms at the
client after each operation. That will ensure the client application
cant issue two operations the same millisecond. It does not help you
if the server time is changed / corrected against a time server as all
Windows machines from XP onwards do by default.

Btw, 1ms = one thousandth of a second.