One of the problems we’ve had for a long time is a runtime error in our application written in XBASE when it tries to close a file on the database server in an NT 4.0 peer to peer network. Network is running NETBEUI at 100MB through CAT 5 cable and a 100MB hub. The databases are stored as .dbf’s and can have as many as 200 open at the same time. We haven’t had much luck fixing this in the code because it seems like the error is an NT error where it occasionally drops a mapped drive. In the
code, back in September, I put in some workaround api calls to detect for lost drive mappings and if detected, our app will attempt to restore the lost drive mappings before continuing (getting around the runtime error.) This is sometimes successful, but most times not. Is there any way to prevent NT from losing drive mappings, or at least making them less frequent? We can’t switch to UNC style naming because the code doesn’t support it fully yet.
