For a start it violates the definition of a IP in basic TCP/IP
standards.
Second, because of this, it also works bad together with many aspects of
TCP/IP where TCP/IP (and IP in general) assumes that the communication
is following TCP/IP specifications...
Despite this, it works OK for most people, but not all.
Areas of conflict:
* IP and fragmented packets
* ICMP traffic related to TCP/IP
* Certain TCP/IP extensions/options
Also, the sad truth is that if the hack of transparent redirection of
TCP/IP traffic hadn't been invented we most likely would have had good
support for proxy discovery in all browsers since several years back..
-- Henrik Nordstrom Squid Hacker McVay, Daniel (DMH) wrote: > > Adam- > > Could you share your logic as to why this is recommended > against? I've been unable to find any documentation or anything > in the FAQs regarding the negatives of employing this technique. > > And would this be set up on a box between the router and the > internal network with 0.0.0.0 pass-through? > > -Dan > DMH LAN AdministratorReceived on Tue Jun 12 2001 - 15:58:43 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:00:43 MST