Re: New issue 16: No packets from other end?

-----BEGIN PGP SIGNED MESSAGE-----


>>>>> "Vijay" == Vijay Devarapalli <[email protected]> writes:
    James> With TCP, the session is bound to the IP address, so if the
    James> address fails, then the session must be terminated, by the
    James> TCP semantics. Having MOBIKE try to change this would require
    James> some kind of cross-layer violation to tell TCP to change.

    >> Typically, in RW scenarioes, an IP address is "extruded" through
    >> the IPsec tunnel, and is bound to a virtual address on the end
    >> host.  Think of this as being like MobileIP's Home-Address.  So,
    >> if you fix the tunnel, then the stuff inside continues unchanged.

    Vijay> MOBIKE should be used to fix the tunnel, but, in IMHO, MOBIKE
    Vijay> shouldnt decide when to fix the tunnel or what address to use
    Vijay> to fix the tunnel.  the IP stack (which might include
    Vijay> MIP6/DNA) should tell the MOBIKE protocol what IP address to
    Vijay> use to fix the tunnel.

  That's fine.
  But the hint that address (A) works while (B) is now broken is an
attribute of the *outside* of the tunnel.

- --
]     "Elmo went to the wrong fundraiser" - The Simpson         |  firewalls  [
]   Michael Richardson,    Xelerance Corporation, Ottawa, ON    |net architect[
] [email protected]      http://www.sandelman.ottawa.on.ca/mcr/ |device driver[
] panic("Just another Debian GNU/Linux using, kernel hacking, security guy"); [
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)
Comment: Finger me for keys

iQCVAwUBQWtLyYqHRg3pndX9AQG/WQQAhDmO46xK8vipO9Jni19gouM5wzXIKZXF
/RB8uC06v0KXqbPjz9ys1oTNZFTR05CKdULW48FegK7yymkAlgXAWD0/c0Ngqivr
5r90Y2ZbCpVqu/bhrXP2E5cwWwVR2TpJXciL1/VAG+KYyyZ0oMvQQpKdn+XPPOq7
qcJ8euf9aUA=
=1Ltp
-----END PGP SIGNATURE-----