Date: Mon, 29 Nov 2004 19:15:29 +0300 From: Gleb Smirnoff <glebius@freebsd.org> To: Andre Oppermann <andre@freebsd.org> Cc: James <haesu@towardex.com> Subject: Re: route cacheing for gif(4) should be optional Message-ID: <20041129161529.GA4770@cell.sick.ru> In-Reply-To: <41AB3CB9.598BB2FB@freebsd.org> References: <20041125140641.GA78210@cell.sick.ru> <20041126025510.GA44246@scylla.towardex.com> <20041126091316.GA84369@cell.sick.ru> <41AB3CB9.598BB2FB@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Nov 29, 2004 at 04:14:01PM +0100, Andre Oppermann wrote: A> > On Thu, Nov 25, 2004 at 09:55:10PM -0500, James wrote: A> > J> On Thu, Nov 25, 2004 at 05:06:41PM +0300, Gleb Smirnoff wrote: A> > J> > Back to this problem: A> > J> > A> > J> > http://freebsd.rambler.ru/bsdmail/freebsd-net_2004/msg01305.html A> > J> > A> > J> > I've found two more people who dislike this feature of gif(4). A> > J> > So I'd like to make it optional. A> > J> > A> > J> > We already have LINK2 flag removing sourceroute filter from gif(4), A> > J> > which is commonly used in asymmetrically routed networks. I suggest A> > J> > to use this flag also for disabling route cacheing, since asymmetricity A> > J> > often appears in dynamically routed networks, and if one runs dynamic A> > J> > routing, he probably wants to remove route cacheing, too. A> > J> A> > J> I'd think we should create a separate option for removing the route A> > J> cache. Sometimes, certain people want to use the tunnel at the highest A> > J> maximum performance possible with both sourceroute filter disabled A> > J> and tunneling routes allocated at their creation time. Perhaps link3 is a A> > J> good place for this option? A> > A> > There is no LINK3 flag :) A> > A> > However, gif(4) does not use LINK0 flag. It was used in past. We can utilize A> > it now. Any objections? A> A> IMO you should scrap it altogether. Well, we have a spare flag at the moment. And splitting flags separately will make one of our users happier. Let us make it separate. Ok? A> However there have been reasons for A> storing the rtentry pointer in struct gif. In the old days ip_output() A> required an rtentry pointer to be passed on, this is no longer the case. A> And it was sort of a safe-guard to make it harder to send the gif encapsulated A> packets back through the same gif interface. That didn't work really well A> and as I say it should be scapped instead of rigged on somewhere else with A> yet another obscure option. ;) As soon as I make route cacheing optional, I'd like to make it off by default. Let me explain: FreeBSD is stable by default, not fast. Routecacheing is not stable. When a route flap occurs in dynamicly routed network my gif tunnels are stuck. I'll describe in manpage, that more performance can be achieved by enabling this route cacheing. Any objections on this default? -- Totus tuus, Glebius. GLEBIUS-RIPN GLEB-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041129161529.GA4770>