Re: TCP end-to-end Semantics

From: Anil Agarwal ([email protected])
Date: Wed Jan 10 2001 - 19:32:31 EST

  • Next message: Leong Kit Hoong, John: "REMOVE"

    In message <[email protected]>,
    William Ivancic wrote:

    >Regarding spoofing:
    >
    >These questions regarding spoofing surfaced today in a meeting. I think I
    >understand what will happen, but I would like your opinion.
    >
    >1. If I am running an application that use TCP and using a Virtual Private
    >Network, will spoofing improve my performance?

    I think the answer is NO. The spoofer will not recognize such packets
    as TCP packets and simply route them.

    One could come up with some clever way to spoof such connections, but
    if the VPN services include IPSec, then I think there is no hope.

    >2. Can a spoofer be utilized in a mobile IP environment were to mobile host
    >or mobile network switches to another network such as in a aeronautical
    >internet?
    >

    I do not know enough about mobile IP to answer that question; but here is
    my best guess.

    If the spoofer is located somewhere near the current base station,
    which changes as the mobile changes cell locations, and IP packets
    are routed from the destination via the current base station, then
    spoofing will probably not work.

    If the spoofer is located somewhere near the "home" base station,
    and if all packets get routed via the "home" base station, then spoofing
    can be made to work.

    Regards,
    Anil Agarwal
    LMGT



    This archive was generated by hypermail 2b29 : Wed Jan 10 2001 - 20:18:07 EST