Jan Engelhardt wrote: > > On Sunday 2008-08-24 20:31, David Miller wrote: > > > >> It appears that the routing code selects the outgoing source address to > >> use for packets is chosen when the socket is established instead of > >> at routing time. > >> > >> The following presents a test case for "unexpected" (from a user's > >> perspective) behavior. > >> > >> Is there any way to make it behave as a user would expect? > > > >No matter when we had made the routing lookup, we would have ended > >up with what you see the kernel doing. > > > >The route is looked up long before netfilter even sees the packets. > > > >The source address selection at the socket level can only "see" the > >original destination address and therefore makes the source address > >selection using that original destination address. > > That's what I thought. With IPv4, one can cheat and use > SNAT/MASQUERADE, but what about IPv6? (Other than using the RAWNAT > extension, which bypasses netfilter connection tracking.) Just for records, I opened a bug report (a feature request) about this at http://bugzilla.kernel.org/show_bug.cgi?id=9767 -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
