index  prev  next

Forced ASCII/UNICODE

--------------------------------------------------------------------------------------

Despite the fact that the client and server negotiate whether to use ASCII or UNICODE, there are lots of places in the protocol where the string is always either one or the other, ignoring the negotiation. Even worse, there are clients (particularly Win2K) that ignore the negotiation for large parts of the protocol.

This latter problem is one of the main reasons for converting your server to negotiate UNICODE, even if you only care about English. Trying to work out if a particular bag of bytes in a packet is actually unicode due to a bug in the client is very error prone. These problems don't happen if you negotiate UNICODE.

--------------------------------------------------------------------------------------

CIFS2001 Seattle
tridge@valinux.com