freebsd-dev/share/man/man4/ng_vlan_rotate.4
Lutz Donnerhacke cfd6422a52 netgraph/ng_vlan_rotate: IEEE 802.1ad VLAN manipulation netgraph type
This node is part of an A10-NSP (L2-BSA) development.

Carrier networks tend to stack three or more tags for internal
purposes and therefore hiding the service tags deep inside of the
stack. When decomposing such an access network frame, the processing
order is typically reversed: First distinguish by service, than by
other means.

This new netgragh node allows to bring the relevant VLAN in front (to
the out-most position). This way other netgraph nodes (like ng_vlan)
can operate on this specific type.

Reviewed by:	manpages (gbe), brueffer (manpages), kp
Approved by:	kp (mentor)
MFC after:	1 month
Relnotes:	yes
Sponsored by:	IKS Service GmbH
Differential Revision: https://reviews.freebsd.org/D22076
2021-01-26 16:53:24 +01:00

253 lines
7.3 KiB
Groff

.\"-
.\" SPDX-License-Identifier: BSD-2-Clause-FreeBSD
.\"
.\" Copyright (c) 2019-2021 IKS Service GmbH
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" Author: Lutz Donnerhacke <lutz@donnerhacke.de>
.\"
.\" $FreeBSD$
.\"
.Dd January 26, 2021
.Dt NG_VLAN_ROTATE 4
.Os
.Sh NAME
.Nm ng_vlan_rotate
.Nd IEEE 802.1ad VLAN manipulation netgraph node type
.Sh SYNOPSIS
.In sys/types.h
.In netgraph.h
.In netgraph/ng_vlan_rotate.h
.Sh DESCRIPTION
The
.Nm
node type manipulates the order of VLAN tags of frames tagged
according to the IEEE 802.1ad (an extension of IEEE 802.1Q) standard
between different hooks.
.Pp
Each node has four special hooks,
.Va original ,
.Va ordered ,
.Va excessive ,
and
.Va incomplete .
.Pp
A frame tagged with an arbitrary number of
.Dv ETHERTYPE_VLAN ,
.Dv ETHERTYPE_QINQ ,
and
.Dv 0x9100
tags received on the
.Va original
hook will be rearranged to a new order of those tags and is sent out
the
.Dq ordered
hook.
After successful processing the
.Va histogram
counter for the observed stack size increments.
.Pp
If it contains fewer VLANs in the stack than the configured
.Va min
limit, the frame is sent out to the
.Va incomplete
hook and the
.Va incomplete
counter increments.
.Pp
If there are more VLANs in the stack than the configured
.Va max
limit, the frame is sent out to the
.Va excessive
hook and the
.Va excessive
counter increments.
.Pp
If the destination hook is not connected, the frame is dropped and the
.Va drops
counter increments.
.Pp
For Ethernet frames received on the
.Va ordered
hook, the transformation is reversed and is passed to the
.Va original
hook.
Please note that this process is identical to the one described
above, besides the ordered/original hooks are swapped and the
transformation is reversed.
.Pp
An Ethernet frame received on the
.Va incomplete
or
.Va excessive
hook is forwarded to the
.Va original
hook without any modification.
.Pp
This node supports only one operation at the moment: Rotation of the
VLANs in the stack.
Setting the configuration parameter
.Va rot
to a positive value, the stack will roll up by this amount.
Negative values will roll down.
A typical scenario is setting the value to 1 in order to bring the
innermost VLAN tag to the outmost level.
Rotation includes the VLAN id, the ether type, and the QOS parameters
pcp and cfi.
Typical QOS handling refers to the outmost setting, so be careful to
keep your QOS intact.
.Sh HOOKS
This node type supports the following hooks:
.Bl -tag -width incomplete
.It Va original
Typically this hook would be connected to a
.Xr ng_ether 4
node, using the
.Va lower
hook connected to a carrier network.
.It Va ordered
Typically this hook would be connected to a
.Xr ng_vlan 4
type node using the
.Va downstream
hook in order to separate services.
.It Va excessive
see below.
.It Va incomplete
Typically those hooks would be attached to a
.Xr ng_eiface 4
type node using the
.Va ether
hook for anomaly monitoring purposes.
.El
.Sh CONTROL MESSAGES
This node type supports the generic control messages, plus the following:
.Bl -tag -width foo
.It Dv NGM_VLANROTATE_GET_CONF Pq Ic getconf
Read the current configuration.
.It Dv NGM_VLANROTATE_SET_CONF Pq Ic setconf
Set the current configuration.
.It Dv NGM_VLANROTATE_GET_STAT Pq Ic getstat
Read the current statistics.
.It Dv NGM_VLANROTATE_CLR_STAT Pq Ic clrstat
Zeroize the statistics.
.It Dv NGM_VLANROTATE_GETCLR_STAT Pq Ic getclrstat
Read the current statistics and zeroize it in one step.
.El
.Sh EXAMPLES
The first example demonstrates how to rotate double or triple tagged
frames so that the innermost C-VLAN can be used as service
discriminator.
The single or double tagged frames (C-VLAN removed) are sent out to an
interface pointing to different infrastucture.
.Bd -literal
#!/bin/sh
BNG_IF=ixl3
VOIP_IF=bge2
ngctl -f- <<EOF
mkpeer ${BNG_IF}: vlan_rotate lower original
name ${BNG_IF}:lower rotate
msg rotate: setconf { min=2 max=3 rot=1 }
mkpeer rotate: vlan ordered downstream
name rotate:ordered services
connect services: ${VOIP_IF} voip lower
msg services: addfilter { vlan=123 hook="voip" }
EOF
.Ed
.Pp
Now inject the following sample frame on the
.Dv BNG_IF
interface:
.Bd -literal
00:00:00:00:01:01 > 00:01:02:03:04:05,
ethertype 802.1Q-9100 (0x9100), length 110: vlan 2, p 1,
ethertype 802.1Q-QinQ, vlan 101, p 0,
ethertype 802.1Q, vlan 123, p 7,
ethertype IPv4, (tos 0x0, ttl 64, id 15994, offset 0, flags [none],
proto ICMP (1), length 84) 192.168.140.101 > 192.168.140.1:
ICMP echo request, id 40234, seq 0, length 64
.Ed
.Pp
The frame ejected on the
.Va ordered
hook will look like this:
.Bd -literal
00:00:00:00:01:01 > 00:01:02:03:04:05,
ethertype 802.1Q (0x8100), length 110: vlan 123, p 7,
ethertype 802.1Q-9100, vlan 2, p 1,
ethertype 802.1Q-QinQ, vlan 101, p 0,
ethertype IPv4, (tos 0x0, ttl 64, id 15994, offset 0, flags [none],
proto ICMP (1), length 84) 192.168.140.101 > 192.168.140.1:
ICMP echo request, id 40234, seq 0, length 64
.Ed
.Pp
Hence, the frame pushed out to the
.Dv VOIP_IF
will have this form:
.Bd -literal
00:00:00:00:01:01 > 00:01:02:03:04:05,
ethertype 802.1Q-9100, vlan 2, p 1,
ethertype 802.1Q-QinQ, vlan 101, p 0,
ethertype IPv4, (tos 0x0, ttl 64, id 15994, offset 0, flags [none],
proto ICMP (1), length 84) 192.168.140.101 > 192.168.140.1:
ICMP echo request, id 40234, seq 0, length 64
.Ed
.Pp
The second example distinguishes between double tagged and single
tagged frames.
.Bd -literal
#!/bin/sh
IN_IF=bge1
ngctl -f- <<EOF
mkpeer ${IN_IF}: vlan_rotate lower original
name ${IN_IF}:lower separate
msg separate: setconf { min=1 max=1 rot=0 }
mkpeer separate: eiface incomplete ether
name separate:incomplete untagged
mkpeer separate: eiface ordered ether
name separate:ordered tagged
EOF
.Ed
.Pp
Setting the
.Va rot
parameter to zero (or omitting it) does not change
the order of the tags within the frame.
Frames with more VLAN tags are dropped.
.Sh SHUTDOWN
This node shuts down upon receipt of a
.Dv NGM_SHUTDOWN
control message, or when all hooks have been disconnected.
.Sh SEE ALSO
.Xr netgraph 4 ,
.Xr ng_eiface 4 ,
.Xr ng_ether 4 ,
.Xr ng_vlan 4 ,
.Xr ngctl 8
.Sh AUTHORS
.An Lutz Donnerhacke Aq Mt lutz@donnerhacke.de