!!!! Automatic configuration with MULTI

Kristian Evensen <kristrev@simula.no> created a daemon which automatically configures the correct routing tables. You can install his tool from %newwin%[[https://github.com/kristrev/multi | github]].

Ondřej Caletka <Ondrej.Caletka@cesnet.cz> created a script for his Gentoo-based system. You can use his script from %newwin%[[https://gist.github.com/oskar456/7264828|github]]. For any questions, please contact him directly.

Dragos Tatulea <dragos@endocode.com> integrated support for Multipath TCP into ConnMan. Checkout his %newwin%[[https://github.com/endocode/connman|github repository]] for access to the source-code

Dragos Tatulea <dragos@endocode.com> integrated support for Multipath TCP into ConnMan. Checkout his %newwin%[[github repository|https://github.com/endocode/connman]] for access to the source-code

Dragos Tatulea <dragos@endocode.com> integrated support for Multipath TCP into ConnMan. Checkout his %newwin%[github repository|https://github.com/endocode/connman] for access to the source-code

Doing the above each time by hand is very cumbersome. Two alternative automatic solutions are available:

!!!! Using the configuration scripts

Kristian Evensen <kristrev@simula.no> created a daemon which automatically configures the correct routing tables. You can install his tool from %newwin%[[https://github.com/kristrev/multi | github]].

!!!! Manual configuration

  * %newwin%[[mptcp_up | https://github.com/multipath-tcp/mptcp-scripts/blob/master/scripts/rt_table/mptcp_up]] - Place it inside /etc/network/if-up.d/ and make it executable.
  * %newwin%[[mptcp_down | https://github.com/multipath-tcp/mptcp-scripts/blob/master/scripts/rt_table/mptcp_down]] - Place it inside /etc/network/if-post-down.d/ and make it executable.

These scripts work most of the time. They use environment variables to configure the routing tables. If they do not work for you, please contact us on the [[mptcp-dev | https://listes-2.sipr.ucl.ac.be/sympa/info/mptcp-dev]] Mailing-List.

Kristian Evensen <kristrev@simula.no> created a daemon which automatically configures the correct routing tables. You can install his tool from %newwin%[[github | https://github.com/kristrev/multi]].

  * %newwin%[[mptcp_up https://github.com/multipath-tcp/mptcp-scripts/blob/master/scripts/rt_table/mptcp_up]] - Place it inside /etc/network/if-up.d/ and make it executable.
  * %newwin%[[mptcp_down https://github.com/multipath-tcp/mptcp-scripts/blob/master/scripts/rt_table/mptcp_down]] - Place it inside /etc/network/if-post-down.d/ and make it executable.

These scripts work most of the time. They use environment variables to configure the routing tables. If they do not work for you, please contact us on the [[mptcp-dev https://listes-2.sipr.ucl.ac.be/sympa/info/mptcp-dev]] Mailing-List.

Kristian Evensen <kristrev@simula.no> created a daemon which automatically configures the correct routing tables. You can install his tool from %newwin%[[github https://github.com/kristrev/multi]].

Many people tell us: Why the hell is my routing working so strangely with multiple addresses? :-) The answer is: '''Any end host having multiple interfaces MUST have one routing table per interface, this is NOT configured automatically (at the moment) by the kernel'''. Here is the reason: We are still (maybe not for long anymore) in an era where end-hosts are assumed to have one default gateway, hence one default outgoing interface. The kernel does support multiple gateways to be defined in the same routing table, but then the first gateway only is used, until it fails, and only then is the second one used. With multiple addresses defined on several interfaces, you want to be able to tell your kernel ''"If I select such source address, please use that specific interface+gateway, not the default ones"''. You achieve this by configuring one routing table per outgoing interface, each routing table being identified by a number. The route selection process then happens in two phases. First the kernel does a lookup in the policy table (that you need to configure with ''ip rules''). The policies, in our case, will be ''For such '''source''' prefix, go to routing table number x''. Then the corresponding routing table is examined to select the gateway based on the destination address.

Now here is concretely how it goes. You have to enable the following compiler-option: ''Networking options -> IP: advanced router'' and ''Networking options -> IP: policy routing''.

Then, you need to configure several routing tables in the following manner:

  * IP-Address: 10.1.1.2
  * Subnet-Mask: 255.255.255.0
  * Gateway: 10.1.1.1

  * IP-Address: 10.1.2.2
  * Subnet-Mask: 255.255.255.0
  * Gateway: 10.1.2.1

The necessary commands are: [[BR]]
{{{
# This creates two different routing tables, that we use based on the source-address.
ip rule add from 10.1.1.2 table 1
ip rule add from 10.1.2.2 table 2

# Configure the two different routing tables
ip route add 10.1.1.0/24 dev eth0 scope link table 1
ip route add default via 10.1.1.1 dev eth0 table 1

ip route add 10.1.2.0/24 dev eth1 scope link table 2
ip route add default via 10.1.2.1 dev eth1 table 2

# default route for the selection process of normal internet-traffic
ip route add default scope global nexthop via 10.1.1.1 dev eth0
}}}

{{{
mptcp-kernel:~# ip rule show
0:      from all lookup local
32764:  from 10.1.2.2 lookup 2
32765:  from 10.1.1.2 lookup 1
32766:  from all lookup main
32767:  from all lookup default

mptcp-kernel:~# ip route
10.1.1.0/24 dev eth0  proto kernel  scope link  src 10.1.1.2
10.1.2.0/24 dev eth1  proto kernel  scope link  src 10.1.2.2
default via 10.1.1.1 dev eth0

mptcp-kernel:~# ip route show table 1
10.1.1.0/24 dev eth0  scope link
default via 10.1.1.1 dev eth0

mptcp-kernel:~# ip route show table 2
10.1.2.0/24 dev eth1  scope link
default via 10.1.2.1 dev eth1 
}}}

Many people tell us: Why the hell is my routing working so strangely with multiple addresses ? :-) The answer is: '''Any end host having multiple interfaces MUST have one routing table per interface, this is NOT configured automatically (at the moment) by the kernel'''. Here is the reason: We are still (maybe not for long anymore) in an era where end-hosts are assumed to have one default gateway, hence one default outgoing interface. The kernel does support multiple gateways to be defined in the same routing table, but then the first gateway only is used, until it fails, and only then is the second one used. With multiple addresses defined on several interfaces, you want to be able to tell your kernel ''If I select such source address, please use that specific interface+gateway, not the default ones''. You achieve this by configuring one routing table per outgoing interface, each routing table being identified by a number. The route selection process then happens in two phases. First the kernel does a lookup in the policy table (that you need to configure with ''ip rules''). The policies, in our case, will be ''For such '''source''' prefix, go to routing table number x''. Then the corresponding routing table is examined to select the gateway based on the destination address.