Yes, and No ;)



Doug, I too send back multiple "types of message" ie.. OK for ok and other mesages for different types of failures but in this case because the server can take a long time to process and in that time could crash the "POOR USERS" could stuck watching time pass them bye while in reality the server process crashed..

I'm looking at what the "com return error" function might give but a multi-thread function, like JAVA or C++ COM object might work here because CODEBASE 6.4 could be used to create and index the incomming files...

The best thing about VFP is there is always SOME way to do it...

Because I have west-wind connection enterprise I also thought of running, out of process servers and giving them a themometer interface so the users could look at the server and SEE their job running. Because of the pooling allowed by WWC ,in this instance, it mimics MTS and would absolutly solve my problem but not exactly a M$ solution :).

BTW thanks for the great ideas and fast response.


JohnP


>>PMFJI,
>>I'm sure this can be done but, wouldn't it defeat the purpose of having a distributed application? I mean, both server and client doing work on the same process at the same time. And it would not help network traffic either.
>
>*rof'l*
>
>Exactly right but that's not what John powers was looking to do. You are correct though in that this violates the whole notion of what we're trying to do.
>
>John? *g*
>
>Best,