When using a very slow nameserver for testing, I've discovered that the current
codebase creates
the socket for the forwarded connection -before- doing the DNS lookup. This is
bad for a variety of
reasons, including:
* tying up unbound sockets for hosts which may not exist;
* making the transport layer decision (ie, IPv4!) before the DNS reply has been
parsed;
* it seems ugly.
I'm not sure if this is the best way of doing this..
Original issue reported on code.google.com by adrian.c...@gmail.com on 4 Apr 2009 at 2:31
Original issue reported on code.google.com by
adrian.c...@gmail.com
on 4 Apr 2009 at 2:31