Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20191021200948.23775-1-mcroce@redhat.com> <20191021200948.23775-5-mcroce@redhat.com>
 <20191023100132.GD8732@netronome.com>
In-Reply-To: <20191023100132.GD8732@netronome.com>
From:   Matteo Croce <mcroce@redhat.com>
Date:   Wed, 23 Oct 2019 18:58:16 +0200
Message-ID: <CAGnkfhy1rsm0Dp_jsuHhfXY0kzMc_hShYmYSX=X8=q-HMtNczg@mail.gmail.com>
Subject: Re: [PATCH net-next 4/4] bonding: balance ICMP echoes in layer3+4 mode
To:     Simon Horman <simon.horman@netronome.com>
Cc:     netdev <netdev@vger.kernel.org>,
        Jay Vosburgh <j.vosburgh@gmail.com>,
        Veaceslav Falico <vfalico@gmail.com>,
        Andy Gospodarek <andy@greyhouse.net>,
        "David S . Miller" <davem@davemloft.net>,
        Stanislav Fomichev <sdf@google.com>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Song Liu <songliubraving@fb.com>,
        Alexei Starovoitov <ast@kernel.org>,
        Paul Blakey <paulb@mellanox.com>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Oct 23, 2019 at 12:01 PM Simon Horman
<simon.horman@netronome.com> wrote:
>
> On Mon, Oct 21, 2019 at 10:09:48PM +0200, Matteo Croce wrote:
> > The bonding uses the L4 ports to balance flows between slaves.
> > As the ICMP protocol has no ports, those packets are sent all to the
> > same device:
> >
> >     # tcpdump -qltnni veth0 ip |sed 's/^/0: /' &
> >     # tcpdump -qltnni veth1 ip |sed 's/^/1: /' &
> >     # ping -qc1 192.168.0.2
> >     1: IP 192.168.0.1 > 192.168.0.2: ICMP echo request, id 315, seq 1, =
length 64
> >     1: IP 192.168.0.2 > 192.168.0.1: ICMP echo reply, id 315, seq 1, le=
ngth 64
> >     # ping -qc1 192.168.0.2
> >     1: IP 192.168.0.1 > 192.168.0.2: ICMP echo request, id 316, seq 1, =
length 64
> >     1: IP 192.168.0.2 > 192.168.0.1: ICMP echo reply, id 316, seq 1, le=
ngth 64
> >     # ping -qc1 192.168.0.2
> >     1: IP 192.168.0.1 > 192.168.0.2: ICMP echo request, id 317, seq 1, =
length 64
> >     1: IP 192.168.0.2 > 192.168.0.1: ICMP echo reply, id 317, seq 1, le=
ngth 64
> >
> > But some ICMP packets have an Identifier field which is
> > used to match packets within sessions, let's use this value in the hash
> > function to balance these packets between bond slaves:
> >
> >     # ping -qc1 192.168.0.2
> >     0: IP 192.168.0.1 > 192.168.0.2: ICMP echo request, id 303, seq 1, =
length 64
> >     0: IP 192.168.0.2 > 192.168.0.1: ICMP echo reply, id 303, seq 1, le=
ngth 64
> >     # ping -qc1 192.168.0.2
> >     1: IP 192.168.0.1 > 192.168.0.2: ICMP echo request, id 304, seq 1, =
length 64
> >     1: IP 192.168.0.2 > 192.168.0.1: ICMP echo reply, id 304, seq 1, le=
ngth 64
> >
> > Signed-off-by: Matteo Croce <mcroce@redhat.com>
>
> I see where this patch is going but it is unclear to me what problem it i=
s
> solving. I would expect ICMP traffic to be low volume and thus able to be
> handled by a single lower-device of a bond.
>
> ...

Hi,

The problem is not balancing the volume, even if it could increase due
to IoT devices pinging some well known DNS servers to check for
connection.
If a bonding slave is down, people using pings to check for
connectivity could fail to detect a broken link if all the packets are
sent to the alive link.
Anyway, although I didn't measure it, the computational overhead of
this changeset should be minimal, and only affect ICMP packets when
the ICMP dissector is used.

Regards,
--=20
Matteo Croce
per aspera ad upstream