1 Virtual Routing and Forwarding (VRF)
2 ====================================
3 The VRF device combined with ip rules provides the ability to create virtual
4 routing and forwarding domains (aka VRFs, VRF-lite to be specific) in the
5 Linux network stack. One use case is the multi-tenancy problem where each
6 tenant has their own unique routing tables and in the very least need
7 different default gateways.
9 Processes can be "VRF aware" by binding a socket to the VRF device. Packets
10 through the socket then use the routing table associated with the VRF
11 device. An important feature of the VRF device implementation is that it
12 impacts only Layer 3 and above so L2 tools (e.g., LLDP) are not affected
13 (ie., they do not need to be run in each VRF). The design also allows
14 the use of higher priority ip rules (Policy Based Routing, PBR) to take
15 precedence over the VRF device rules directing specific traffic as desired.
17 In addition, VRF devices allow VRFs to be nested within namespaces. For
18 example network namespaces provide separation of network interfaces at the
19 device layer, VLANs on the interfaces within a namespace provide L2 separation
20 and then VRF devices provide L3 separation.
24 A VRF device is created with an associated route table. Network interfaces
25 are then enslaved to a VRF device:
27 +-----------------------------+
28 | vrf-blue | ===> route table 10
29 +-----------------------------+
31 +------+ +------+ +-------------+
32 | eth1 | | eth2 | ... | bond1 |
33 +------+ +------+ +-------------+
39 Packets received on an enslaved device and are switched to the VRF device
40 in the IPv4 and IPv6 processing stacks giving the impression that packets
41 flow through the VRF device. Similarly on egress routing rules are used to
42 send packets to the VRF device driver before getting sent out the actual
43 interface. This allows tcpdump on a VRF device to capture all packets into
44 and out of the VRF as a whole.[1] Similarly, netfilter[2] and tc rules can be
45 applied using the VRF device to specify rules that apply to the VRF domain
48 [1] Packets in the forwarded state do not flow through the device, so those
49 packets are not seen by tcpdump. Will revisit this limitation in a
52 [2] Iptables on ingress supports PREROUTING with skb->dev set to the real
53 ingress device and both INPUT and PREROUTING rules with skb->dev set to
54 the VRF device. For egress POSTROUTING and OUTPUT rules can be written
55 using either the VRF device or real egress device.
59 1. VRF device is created with an association to a FIB table.
60 e.g, ip link add vrf-blue type vrf table 10
61 ip link set dev vrf-blue up
63 2. An l3mdev FIB rule directs lookups to the table associated with the device.
64 A single l3mdev rule is sufficient for all VRFs. The VRF device adds the
65 l3mdev rule for IPv4 and IPv6 when the first device is created with a
66 default preference of 1000. Users may delete the rule if desired and add
67 with a different priority or install per-VRF rules.
69 Prior to the v4.8 kernel iif and oif rules are needed for each VRF device:
70 ip ru add oif vrf-blue table 10
71 ip ru add iif vrf-blue table 10
73 3. Set the default route for the table (and hence default route for the VRF).
74 ip route add table 10 unreachable default metric 4278198272
76 This high metric value ensures that the default unreachable route can
77 be overridden by a routing protocol suite. FRRouting interprets
78 kernel metrics as a combined admin distance (upper byte) and priority
79 (lower 3 bytes). Thus the above metric translates to [255/8192].
81 4. Enslave L3 interfaces to a VRF device.
82 ip link set dev eth1 master vrf-blue
84 Local and connected routes for enslaved devices are automatically moved to
85 the table associated with VRF device. Any additional routes depending on
86 the enslaved device are dropped and will need to be reinserted to the VRF
87 FIB table following the enslavement.
89 The IPv6 sysctl option keep_addr_on_down can be enabled to keep IPv6 global
90 addresses as VRF enslavement changes.
91 sysctl -w net.ipv6.conf.all.keep_addr_on_down=1
93 5. Additional VRF routes are added to associated table.
94 ip route add table 10 ...
99 Applications that are to work within a VRF need to bind their socket to the
102 setsockopt(sd, SOL_SOCKET, SO_BINDTODEVICE, dev, strlen(dev)+1);
104 or to specify the output device using cmsg and IP_PKTINFO.
106 TCP & UDP services running in the default VRF context (ie., not bound
107 to any VRF device) can work across all VRF domains by enabling the
108 tcp_l3mdev_accept and udp_l3mdev_accept sysctl options:
109 sysctl -w net.ipv4.tcp_l3mdev_accept=1
110 sysctl -w net.ipv4.udp_l3mdev_accept=1
112 netfilter rules on the VRF device can be used to limit access to services
113 running in the default VRF context as well.
115 The default VRF does not have limited scope with respect to port bindings.
116 That is, if a process does a wildcard bind to a port in the default VRF it
117 owns the port across all VRF domains within the network namespace.
119 ################################################################################
121 Using iproute2 for VRFs
122 =======================
123 iproute2 supports the vrf keyword as of v4.7. For backwards compatibility this
124 section lists both commands where appropriate -- with the vrf keyword and the
125 older form without it.
129 To instantiate a VRF device and associate it with a table:
130 $ ip link add dev NAME type vrf table ID
132 As of v4.8 the kernel supports the l3mdev FIB rule where a single rule
133 covers all VRFs. The l3mdev rule is created for IPv4 and IPv6 on first
138 To list VRFs that have been created:
139 $ ip [-d] link show type vrf
140 NOTE: The -d option is needed to show the table id
143 $ ip -d link show type vrf
144 11: mgmt: <NOARP,MASTER,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
145 link/ether 72:b3:ba:91:e2:24 brd ff:ff:ff:ff:ff:ff promiscuity 0
146 vrf table 1 addrgenmode eui64
147 12: red: <NOARP,MASTER,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
148 link/ether b6:6f:6e:f6:da:73 brd ff:ff:ff:ff:ff:ff promiscuity 0
149 vrf table 10 addrgenmode eui64
150 13: blue: <NOARP,MASTER,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
151 link/ether 36:62:e8:7d:bb:8c brd ff:ff:ff:ff:ff:ff promiscuity 0
152 vrf table 66 addrgenmode eui64
153 14: green: <NOARP,MASTER,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
154 link/ether e6:28:b8:63:70:bb brd ff:ff:ff:ff:ff:ff promiscuity 0
155 vrf table 81 addrgenmode eui64
160 $ ip -br link show type vrf
161 mgmt UP 72:b3:ba:91:e2:24 <NOARP,MASTER,UP,LOWER_UP>
162 red UP b6:6f:6e:f6:da:73 <NOARP,MASTER,UP,LOWER_UP>
163 blue UP 36:62:e8:7d:bb:8c <NOARP,MASTER,UP,LOWER_UP>
164 green UP e6:28:b8:63:70:bb <NOARP,MASTER,UP,LOWER_UP>
167 3. Assign a Network Interface to a VRF
169 Network interfaces are assigned to a VRF by enslaving the netdevice to a
171 $ ip link set dev NAME master NAME
173 On enslavement connected and local routes are automatically moved to the
174 table associated with the VRF device.
177 $ ip link set dev eth0 master mgmt
180 4. Show Devices Assigned to a VRF
182 To show devices that have been assigned to a specific VRF add the master
183 option to the ip command:
184 $ ip link show vrf NAME
185 $ ip link show master NAME
188 $ ip link show vrf red
189 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP mode DEFAULT group default qlen 1000
190 link/ether 02:00:00:00:02:02 brd ff:ff:ff:ff:ff:ff
191 4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP mode DEFAULT group default qlen 1000
192 link/ether 02:00:00:00:02:03 brd ff:ff:ff:ff:ff:ff
193 7: eth5: <BROADCAST,MULTICAST> mtu 1500 qdisc noop master red state DOWN mode DEFAULT group default qlen 1000
194 link/ether 02:00:00:00:02:06 brd ff:ff:ff:ff:ff:ff
197 Or using the brief output:
198 $ ip -br link show vrf red
199 eth1 UP 02:00:00:00:02:02 <BROADCAST,MULTICAST,UP,LOWER_UP>
200 eth2 UP 02:00:00:00:02:03 <BROADCAST,MULTICAST,UP,LOWER_UP>
201 eth5 DOWN 02:00:00:00:02:06 <BROADCAST,MULTICAST>
204 5. Show Neighbor Entries for a VRF
206 To list neighbor entries associated with devices enslaved to a VRF device
207 add the master option to the ip command:
208 $ ip [-6] neigh show vrf NAME
209 $ ip [-6] neigh show master NAME
212 $ ip neigh show vrf red
213 10.2.1.254 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE
214 10.2.2.254 dev eth2 lladdr 5e:54:01:6a:ee:80 REACHABLE
216 $ ip -6 neigh show vrf red
217 2002:1::64 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE
220 6. Show Addresses for a VRF
222 To show addresses for interfaces associated with a VRF add the master
223 option to the ip command:
224 $ ip addr show vrf NAME
225 $ ip addr show master NAME
228 $ ip addr show vrf red
229 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000
230 link/ether 02:00:00:00:02:02 brd ff:ff:ff:ff:ff:ff
231 inet 10.2.1.2/24 brd 10.2.1.255 scope global eth1
232 valid_lft forever preferred_lft forever
233 inet6 2002:1::2/120 scope global
234 valid_lft forever preferred_lft forever
235 inet6 fe80::ff:fe00:202/64 scope link
236 valid_lft forever preferred_lft forever
237 4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000
238 link/ether 02:00:00:00:02:03 brd ff:ff:ff:ff:ff:ff
239 inet 10.2.2.2/24 brd 10.2.2.255 scope global eth2
240 valid_lft forever preferred_lft forever
241 inet6 2002:2::2/120 scope global
242 valid_lft forever preferred_lft forever
243 inet6 fe80::ff:fe00:203/64 scope link
244 valid_lft forever preferred_lft forever
245 7: eth5: <BROADCAST,MULTICAST> mtu 1500 qdisc noop master red state DOWN group default qlen 1000
246 link/ether 02:00:00:00:02:06 brd ff:ff:ff:ff:ff:ff
249 $ ip -br addr show vrf red
250 eth1 UP 10.2.1.2/24 2002:1::2/120 fe80::ff:fe00:202/64
251 eth2 UP 10.2.2.2/24 2002:2::2/120 fe80::ff:fe00:203/64
255 7. Show Routes for a VRF
257 To show routes for a VRF use the ip command to display the table associated
259 $ ip [-6] route show vrf NAME
260 $ ip [-6] route show table ID
263 $ ip route show vrf red
264 unreachable default metric 4278198272
265 broadcast 10.2.1.0 dev eth1 proto kernel scope link src 10.2.1.2
266 10.2.1.0/24 dev eth1 proto kernel scope link src 10.2.1.2
267 local 10.2.1.2 dev eth1 proto kernel scope host src 10.2.1.2
268 broadcast 10.2.1.255 dev eth1 proto kernel scope link src 10.2.1.2
269 broadcast 10.2.2.0 dev eth2 proto kernel scope link src 10.2.2.2
270 10.2.2.0/24 dev eth2 proto kernel scope link src 10.2.2.2
271 local 10.2.2.2 dev eth2 proto kernel scope host src 10.2.2.2
272 broadcast 10.2.2.255 dev eth2 proto kernel scope link src 10.2.2.2
274 $ ip -6 route show vrf red
275 local 2002:1:: dev lo proto none metric 0 pref medium
276 local 2002:1::2 dev lo proto none metric 0 pref medium
277 2002:1::/120 dev eth1 proto kernel metric 256 pref medium
278 local 2002:2:: dev lo proto none metric 0 pref medium
279 local 2002:2::2 dev lo proto none metric 0 pref medium
280 2002:2::/120 dev eth2 proto kernel metric 256 pref medium
281 local fe80:: dev lo proto none metric 0 pref medium
282 local fe80:: dev lo proto none metric 0 pref medium
283 local fe80::ff:fe00:202 dev lo proto none metric 0 pref medium
284 local fe80::ff:fe00:203 dev lo proto none metric 0 pref medium
285 fe80::/64 dev eth1 proto kernel metric 256 pref medium
286 fe80::/64 dev eth2 proto kernel metric 256 pref medium
287 ff00::/8 dev red metric 256 pref medium
288 ff00::/8 dev eth1 metric 256 pref medium
289 ff00::/8 dev eth2 metric 256 pref medium
290 unreachable default dev lo metric 4278198272 error -101 pref medium
292 8. Route Lookup for a VRF
294 A test route lookup can be done for a VRF:
295 $ ip [-6] route get vrf NAME ADDRESS
296 $ ip [-6] route get oif NAME ADDRESS
299 $ ip route get 10.2.1.40 vrf red
300 10.2.1.40 dev eth1 table red src 10.2.1.2
303 $ ip -6 route get 2002:1::32 vrf red
304 2002:1::32 from :: dev eth1 table red proto kernel src 2002:1::2 metric 256 pref medium
307 9. Removing Network Interface from a VRF
309 Network interfaces are removed from a VRF by breaking the enslavement to
311 $ ip link set dev NAME nomaster
313 Connected routes are moved back to the default table and local entries are
314 moved to the local table.
317 $ ip link set dev eth0 nomaster
319 --------------------------------------------------------------------------------
321 Commands used in this example:
323 cat >> /etc/iproute2/rt_tables.d/vrf.conf <<EOF
336 ip link add ${VRF} type vrf table ${TBID}
338 if [ "${VRF}" != "mgmt" ]; then
339 ip route add table ${TBID} unreachable default metric 4278198272
341 ip link set dev ${VRF} up
345 ip link set dev eth0 master mgmt
348 ip link set dev eth1 master red
349 ip link set dev eth2 master red
350 ip link set dev eth5 master red
353 ip link set dev eth3 master blue
356 ip link set dev eth4 master green
359 Interface addresses from /etc/network/interfaces:
361 iface eth0 inet static
363 netmask 255.255.255.0
366 iface eth0 inet6 static
371 iface eth1 inet static
373 netmask 255.255.255.0
375 iface eth1 inet6 static
380 iface eth2 inet static
382 netmask 255.255.255.0
384 iface eth2 inet6 static
389 iface eth3 inet static
391 netmask 255.255.255.0
393 iface eth3 inet6 static
398 iface eth4 inet static
400 netmask 255.255.255.0
402 iface eth4 inet6 static