97 lines
4.0 KiB
Plaintext
97 lines
4.0 KiB
Plaintext
Virtual Routing and Forwarding (VRF)
|
|
====================================
|
|
The VRF device combined with ip rules provides the ability to create virtual
|
|
routing and forwarding domains (aka VRFs, VRF-lite to be specific) in the
|
|
Linux network stack. One use case is the multi-tenancy problem where each
|
|
tenant has their own unique routing tables and in the very least need
|
|
different default gateways.
|
|
|
|
Processes can be "VRF aware" by binding a socket to the VRF device. Packets
|
|
through the socket then use the routing table associated with the VRF
|
|
device. An important feature of the VRF device implementation is that it
|
|
impacts only Layer 3 and above so L2 tools (e.g., LLDP) are not affected
|
|
(ie., they do not need to be run in each VRF). The design also allows
|
|
the use of higher priority ip rules (Policy Based Routing, PBR) to take
|
|
precedence over the VRF device rules directing specific traffic as desired.
|
|
|
|
In addition, VRF devices allow VRFs to be nested within namespaces. For
|
|
example network namespaces provide separation of network interfaces at L1
|
|
(Layer 1 separation), VLANs on the interfaces within a namespace provide
|
|
L2 separation and then VRF devices provide L3 separation.
|
|
|
|
Design
|
|
------
|
|
A VRF device is created with an associated route table. Network interfaces
|
|
are then enslaved to a VRF device:
|
|
|
|
+-----------------------------+
|
|
| vrf-blue | ===> route table 10
|
|
+-----------------------------+
|
|
| | |
|
|
+------+ +------+ +-------------+
|
|
| eth1 | | eth2 | ... | bond1 |
|
|
+------+ +------+ +-------------+
|
|
| |
|
|
+------+ +------+
|
|
| eth8 | | eth9 |
|
|
+------+ +------+
|
|
|
|
Packets received on an enslaved device and are switched to the VRF device
|
|
using an rx_handler which gives the impression that packets flow through
|
|
the VRF device. Similarly on egress routing rules are used to send packets
|
|
to the VRF device driver before getting sent out the actual interface. This
|
|
allows tcpdump on a VRF device to capture all packets into and out of the
|
|
VRF as a whole.[1] Similiarly, netfilter [2] and tc rules can be applied
|
|
using the VRF device to specify rules that apply to the VRF domain as a whole.
|
|
|
|
[1] Packets in the forwarded state do not flow through the device, so those
|
|
packets are not seen by tcpdump. Will revisit this limitation in a
|
|
future release.
|
|
|
|
[2] Iptables on ingress is limited to NF_INET_PRE_ROUTING only with skb->dev
|
|
set to real ingress device and egress is limited to NF_INET_POST_ROUTING.
|
|
Will revisit this limitation in a future release.
|
|
|
|
|
|
Setup
|
|
-----
|
|
1. VRF device is created with an association to a FIB table.
|
|
e.g, ip link add vrf-blue type vrf table 10
|
|
ip link set dev vrf-blue up
|
|
|
|
2. Rules are added that send lookups to the associated FIB table when the
|
|
iif or oif is the VRF device. e.g.,
|
|
ip ru add oif vrf-blue table 10
|
|
ip ru add iif vrf-blue table 10
|
|
|
|
Set the default route for the table (and hence default route for the VRF).
|
|
e.g, ip route add table 10 prohibit default
|
|
|
|
3. Enslave L3 interfaces to a VRF device.
|
|
e.g, ip link set dev eth1 master vrf-blue
|
|
|
|
Local and connected routes for enslaved devices are automatically moved to
|
|
the table associated with VRF device. Any additional routes depending on
|
|
the enslaved device will need to be reinserted following the enslavement.
|
|
|
|
4. Additional VRF routes are added to associated table.
|
|
e.g., ip route add table 10 ...
|
|
|
|
|
|
Applications
|
|
------------
|
|
Applications that are to work within a VRF need to bind their socket to the
|
|
VRF device:
|
|
|
|
setsockopt(sd, SOL_SOCKET, SO_BINDTODEVICE, dev, strlen(dev)+1);
|
|
|
|
or to specify the output device using cmsg and IP_PKTINFO.
|
|
|
|
|
|
Limitations
|
|
-----------
|
|
VRF device currently only works for IPv4. Support for IPv6 is under development.
|
|
|
|
Index of original ingress interface is not available via cmsg. Will address
|
|
soon.
|