On Sun, Jun 21, 2009 at 5:57 PM, patrick keshishian<pkesh...@gmail.com> wrote: > On Sun, Jun 21, 2009 at 3:42 PM, Philip Guenther<guent...@gmail.com> wrote: >> On Sun, Jun 21, 2009 at 10:36 AM, patrick keshishian<pkesh...@gmail.com> wrote: >>> *aham* ... was this a really stupid question? >> >> Well, you elided useful data by only including part of the netstat >> output, you obfuscated it to make it harder to read, you failed to >> even mention what version of OpenBSD you're running, *and* you >> actually have a solution to your problem. Why should anyone bother to >> answer? > > ouch... but thanks for taking the time to reply. > > well, you have some good points there, but if you read carefully, my > post wasn't of the "Hey everyone please help me!" flavour. It was of > the form "I notice this on openbsd and this on this other platform, I > wonder which is the expected behavior?"
Sure, but how should someone decide that the behavior is expected when you leave out chunks of the information that describes your setup? Do I need to have a multipath + ppp setup to be able to help? > This was noticed on periodically-updated openbsd macppc-snapshots > since pre 4.4 release until one from 2 months ago, which I'm currently > running. So you're now running some undisclosed version of 4.5-current? Wait, does that "until one from 2 months ago" mean that the behavior changed when you most recently updated the snapshot you're running?!? >>> Maybe I just wrote too many words. In simple terms, once a new route >>> has been added to the routing table, all traffic should consider the >>> new route right? So, is the ppp interface treated differently when it >>> comes to routing in OpenBSD? >> >> Does this quote from the netstat(8) manpage explain the behavior? >> Connection oriented protocols normally hold on to a single route >> for the duration of a connection while connectionless protocols obtain a >> route while sending to the same destination. > > ah, yes. this is good, as it confirms part of my observation; note > that i was not specific on the type of socket used, because it did not > make a difference. I simply said "same socket descriptor", indicating > one created prior to the establishment of the new route. Sigh. Was it really too hard to say "this occurs with both TCP and UDP and even straight ICMPs from ping" in your original note? HTH were we supposed to know that *this* time, "packets" really meant all three of those and wasn't just being used as a shorthand for whatever was being run at the moment? You may be 100% accurate in your use of language, but you're posting to a list where many people are more careless about terms, so the default expectation is that unless someone uses precise terms, they probably haven't considered the distinctions. > e.g., I can start a ping going for the particular host on the remote > network, next establish the route and the pings continue out on the > physical interface. If I start a new ping, those packets, now, go > through the ppp0 interface. As verified with tcpdump. > > So, it seems, based on my observations, routes are "sticky" with > respect to sockets; even non-TCP sockets, which seems bit odd. Do you > not agree? Still asking for people to state expectations on zero data. My crystal ball says that that netstat info would have been interesting, but since you apparently only are interested in responses from people that happen to have multipath setups and use ppp, I guess I can't help you. Good luck! Philip Guenther
