Messages in this thread | | | Date | Tue, 20 Oct 1998 14:40:49 +0200 | From | Erik Corry <> | Subject | Re: Dynamic IP hack (PR#294) |
| |
On Tue, Oct 20, 1998 at 01:54:24PM +0100, Alan Cox wrote: > Erik Corry wrote: > > 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. > > And the only person we might screw up is ourselves in that case.
Who else do we screw up in the other case?
> If you simply supress sending with source addresses you dont currently > posess then the kernel is behaving right. Diald can and should be left > to sort the rest out.
Or ipppd or pppd, or even manually operated non-dial-on-demand PPP connections with dynamic addresses. In all those cases you have the problem that the app hangs and isn't told that the socket is dead. Which it is.
> I still believe this hack doesnt belong in 2.1.x
I'm sorry to hear that, especially since you were the one that put it in 2.0. Let me make a prediction that if no better solution appears, the hack will be in SuSE's version of 2.2. I have no connection with SuSE, but they have a lot of ISDN customers, and something like this is very important to them.
-- 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/
| |