Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 15 Aug 2011 10:06:20 -0500
From:      David Duchscher <daved@tamu.edu>
To:        Gleb Smirnoff <glebius@FreeBSD.org>
Cc:        net@FreeBSD.org
Subject:   Re: new CARP implementation
Message-ID:  <E2D7B3D7-27D0-488C-9873-23912A803B0C@tamu.edu>
In-Reply-To: <20110815132825.GJ43567@glebius.int.ru>
References:  <20110810160526.GO43567@FreeBSD.org> <5D7408D3-FAA1-4E22-A136-83DC75D47837@tamu.edu> <20110814084813.GA43567@glebius.int.ru> <67BC462C-0F5D-41E2-B739-CFC9EB417FA6@tamu.edu> <20110815120750.GF43567@glebius.int.ru> <A23AEC0C-6C23-4FC9-B36A-01CF3DE1311F@tamu.edu> <20110815132825.GJ43567@glebius.int.ru>

next in thread | previous in thread | raw e-mail | index | archive | help

--Apple-Mail=_D970B3AE-0BA8-4152-9F6C-140F88DC8920
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=koi8-r

On Aug 15, 2011, at 8:28 AM, Gleb Smirnoff wrote:

> On Mon, Aug 15, 2011 at 07:56:14AM -0500, David Duchscher wrote:
> D> > On Sun, Aug 14, 2011 at 03:56:28PM -0500, David Duchscher wrote:
> D> > D> > On Sat, Aug 13, 2011 at 07:32:06PM -0500, David Duchscher =
wrote:
> D> > D> > D> My two cents.
> D> > D> > D>=20
> D> > D> > D> We rely on the arp load balance feature.  We certainly =
don't find it useless.  Looking at ip load balancing, it would also mean =
that we would no longer be able to grow bandwidth with additional =
systems since all boxes must receive all traffic. I only humbling ask =
that some sort of load balancing feature be included when this goes =
live.
> D> > D> >=20
> D> > D> > Ok, I will make effort to support it. I will inform when =
patch would
> D> > D> > be updated.
> D> > D>=20
> D> > D> Thank you.
> D> >=20
> D> > On closer look it appeared that restoring ARP balancing as it =
was, isn't going
> D> > to be easy. The essence of ARP balancing is that different vhids =
possess the
> D> > same IP address. Converting that to new scheme would mean that =
same IP prefixes
> D> > exist on one interface, which is impossible in current networking =
stack. And
> D> > making it possible would be a bloody hack.
> D> >=20
> D> > So I'd prefer to settle current code a bit, commit it to head, =
after 9.0 is
> D> > forked and released... Test and settle code a bit more... And =
then work on
> D> > ARP and IP balancing. That would probably require bringing in =
some intermediate
> D> > structure along with struct carp_softc, that would group softcs =
into
> D> > balancing groups. That is already done in OpenBSD. Not sure that =
our balancing
> D> > would be compatible with OpenBSD's, however the current is not =
already, since
> D> > OpenBSD changed their hashing function after we merged carp(4) to =
FreeBSD.
> D>=20
> D> This sound good to me.  I have no requirement for compatibility =
with OpenBSD.  In addition, we only use the extended support versions of =
FreeBSD so 9.0 will not be something we will put into production.
>=20
> The new CARP isn't going to appear in 9.x, since I am too late with my =
patch. It
> would be present in 10.0 and later. However, I plan to maintain an =
easily applicable
> patch for 9.x.
>=20
> However, I'd suggest you to try new CARP earlier than 10.1-RELEASE is =
out. May be in
> a test lab. I'm afraid that number of people utilizing ARP balancing =
is so small,
> that without your testing, feature would go untested into release ;)

Already do.  We have over 20k devices that are behind NAT run on top of =
FreeBSD so its rather important we don't break it. :)

Thanks for the update,
--
DaveD


--Apple-Mail=_D970B3AE-0BA8-4152-9F6C-140F88DC8920--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?E2D7B3D7-27D0-488C-9873-23912A803B0C>