Date: Thu, 24 Jun 2004 11:17:27 -0400 From: Sven Willenberger <sven@dmv.com> To: Dan Nelson <dnelson@allantgroup.com> Cc: freebsd-hackers@freebsd.org Subject: Re: using netgraph to connect 2 physical interfaces into one virtual interface Message-ID: <1088090247.8744.7.camel@lanshark.dmv.com> In-Reply-To: <20040624143343.GA56406@dan.emsphone.com> References: <1088030164.29367.57.camel@lanshark.dmv.com> <20040624143343.GA56406@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 2004-06-24 at 09:33 -0500, Dan Nelson wrote: > In the last episode (Jun 23), Sven Willenberger said: > > I am having a lot of trouble trying to make the following work (after > > some exhaustive googling etc) > > > > Goal: 2 interfaces (em0 and em1) to be "combined" or bonded into one > > virtual interface so as to provide both increased throughput and > > failover. Both physical ports connected to either the same or different > > switches with a virtual gateway (the configuration for which is being > > haandled separately). > > > > What I have tried (using netgraph) and the results: > > > > 1) (from the ng_one2many manpage): > > 2) adapted from freebsd-security (derkweiler) http://www.derkeiler.com/Mailing-Lists/FreeBSD-Security/2004-01/0084.html thread : > > > > So my question is, without trying to get into ng_fec (which I understand > > will also need hardware support on the other end -- blades, etc), how > > ng_fec needs just as much hardware support as one2many: the system at > the other end must be able to handle port aggregation, and must be able > to be manually configured. Both nodes do the same thing, in slightly > different ways. > Actually I was wondering if one2many needed the same support that ng_fec needs - and you answered that question, albeit inversely. I did get ng_fec to run (using only three lines) and I don't have that issue of (DUP!) in my packets or any performance hit as long as fec0 is ifconfig'd with arp and the two physical interfaces are on the same or connected switch(es). ngctl mkpeer fec dummy fec (which spits out an error about not being able to name the fec0 node) ngctl msg fec0: add_iface '"em0"' ngctl msg fec0: add_iface '"em1"' ifconfig fec0 up So what would the equivalent be using ng_one2many and how do I get the same throughput and lack of DUP packets that I can achieve with ng_fec. My understanding was that ng_fec was being deprecated in favor of ng_one2many ? Sven
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1088090247.8744.7.camel>