Probably, but at least the socket gets RST the next time
we receive a packet from the peer.
There's no solution here that doesn't violate protocol
requirements. If someone presses the big red button and
we reboot in under 120s we also violate the time wait
requirements of the TCP protocol.
A dynamic-address PPP connection going down is not the
way the Internet was supposed to work, but it's real life.
With Andi's solution we kill our end of the socket without
telling the peer, which can't be conformant.
(Juanjo, who wrote the original dyn_addr hack suggested
we send out 'forged' RST packets to close the other
end down, which would be friendly, but non-conformant.
I like the idea, though I don't know how many users would
want to spend a phone unit on it each time).
At the moment we continuously redial in a desperate attempt
to get our old address back. Since most ISPs have pools
of 100 addresses or more that would be a pointless and
expensive tactic, even if we suppressed sending packets
with an address we no longer have.
-- Erik Corry erik@arbat.com Ceterum censeo, Microsoftem esse delendam!- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/