Permalink
Find file
3c6ca93 Nov 10, 2016
@axn @rubo77 @p4u @mvdan @rbpasker
873 lines (646 sloc) 44.1 KB

BMX6

Bmx6 is a mesh routing protocol for Linux based operating systems. The following intro provides kind of tutorial to get started.

Content

Note: This document is written using Markdown syntax. Modifications should be synced via README.md file in bmx6 repositories bmx6.net and github.com. Nice syntax examples are here.

Installation

Requirements

The following tools are needed to obtain, compile, and install bmx6:

  • git (debian package: git-core)
  • gcc
  • make

The following Linux-kernel modules are needed (depending on used bmx6 features)

  • ipv6
  • tunnel6
  • ip6_tunnel

Downloading

Latest development sources are available from bmx6 git repository:

git clone git://qmp.cat/bmx6.git # alternative: https://github.com/axn/bmx6.git
cd bmx6

Compile and Install

To only compile the main bmx6 daemon (no bmx6 plugins):

make
sudo make install

Installing in OpenWRT

Bmx6 is currently in the official routing feed of OpenWRT, so to install it from a existing system you can use opkg:

opkg install bmx6 bmx6-uci-config

If you are compiling your own OpenWRT, you can add the routing feed (already enabled by default) which can be found here: https://github.com/openwrt-routing/packages

Then run "make menuconfig" and select the bmx6 package in Networking -> Routing and redirection

It is recommended to select also, at least, the uci plugin (bmx6-uci-config)

You can select "luci-app-bmx6" to have a nice web interface for manage and monitorize the routing daemon.

Finally type "make" to build the image.

Installing in ArchLinux

Bmx6 is currently in the AUR repository of Archlinux. The current package includes JSON, SMS and Table plugins. https://aur4.archlinux.org/packages/bmx6/

You can use the yaourt package manager to install it:

yaourt -S bmx6

Once installed you can use systemd to start it:

systemctl start bmx6

Remember to vote the package so it might be included into the officialy community repository of ArchLinux!

Usage (hello mesh)

Starting

In the most simple configuration, the only required parameter are the interfaces names that should be used for meshing. The following example starts bmx6 on interface wlan0:

root@mlc1001:~# bmx6 dev=wlan0

However, to let this simple command work as expected also check the following basic requirements:

  • bmx6 must be executed in root context (with super user permissions). If you are not already root, prepend all commands with sudo (eg: sudo bmx6 dev=eth1 ).

  • No IP address needs to be configured. By default bmx6 assumes IPv6 and autoconfigures a ULA-based IPv6 address for each interface based on the MAC address of the device. The only pre-requisite is that the interfaces must be in the up state, E.G.: ip link set wlan0 up.

    If you are using a wireless interface, the interface settings must have been configured using iwconfig to communicate with bmx6 daemons running on other nodes. This is a typical configuration for a wireless mesh setup:

    iwconfig wlan0 mode ad-hoc ap 02:ca:ff:ee:ba:be channel 11 essid my-mesh-network

  • Bmx6 (by default) works in daemon mode, thus sends itself to background and gives back a prompt. To let it run in foreground specify a debug level with the startup command like:

     bmx6 debug=0 dev=eth1 
    Of course, you may need to kill a previously started bmx6 daemon beforehand (killall bmx6)

If everything went fine bmx6 is running now, searches for neighboring bmx6 daemons via the configured interface (link), and coordinates with them to learn about existence-of and routes-to all other bmx6 nodes in the network.

Monitoring bmx6

To access debug and status information of a running bmx6 daemon, a second bmx6 process can be launched in client mode (with the --connect or -c parameter) to connect to the main bmx6 daemon and retrieve the desired information.

In the following, a few examples will be discussed. Debug levels with different verbosity and scope are accessible with the --debug or -d parameter.

  • Debug level 0 only reports critical events
  • Debug level 3 reports relevant changes and
  • Debug level 4 reports everything.
  • Debug level 12 dump in and outgoing protocol traffic

For example, bmx6 -cd3 runs a bmx6 client process at debug level 3, connected to the main daemon and logs the output to stdout until terminated with ctrl-c.

Status, network, and statistic information are also accessible via their own parameters:

  • status
  • interfaces
  • links
  • originators
  • descriptions, plus optional sub-parameters for filtering
  • tunnels
  • traffic=DEV where DEV:=all, eth1, etc.
root@mlc1001:~# bmx6 -c status
version        compatibility codeVersion globalId                     primaryIp                       myLocalId uptime     cpu nodes
BMX6-0.1-alpha 16            9           mlc1001.7A7422752001EC4AC4C8 fd66:66:66:0:a2cd:efff:fe10:101 24100101  0:00:40:37 0.1 4

So apart from version, compatibility number, and code, the status reveals the daemon's Global ID and Local ID, its primary (self-configured) IPv6 address, the elapsed time since it was started (40 minutes), its current cpu consumption (0.1%), and the total number of learned nodes in the network (4, including itself).

These desired types can be combined. Also the above given example shows kind of shortcut. The long argument would be: bmx6 connect show=status. A more informative case using the long form would be:

root@mlc1001:~# bmx6 connect show=status show=interfaces show=links show=originators show=tunnels
status:
version        compatibility codeVersion globalId                     primaryIp                       myLocalId uptime     cpu nodes
BMX6-0.1-alpha 16            9           mlc1001.7A7422752001EC4AC4C8 fd66:66:66:0:a2cd:efff:fe10:101 06100101  0:00:53:19 0.3 4
interfaces:
devName state type     rateMin rateMax llocalIp                    globalIp                           multicastIp primary
eth1    UP    ethernet 1000M   1000M   fe80::a2cd:efff:fe10:101/64 fd66:66:66:0:a2cd:efff:fe10:101/64 ff02::2     1
links:
globalId                     llocalIp                 viaDev rxRate txRate bestTxLink routes wantsOgms nbLocalId
mlc1000.0AE58311046412F248CD fe80::a2cd:efff:fe10:1   eth1   100    100    1          1      1         9B100001
mlc1002.91DCF042934B5913BB00 fe80::a2cd:efff:fe10:201 eth1   100    100    1          2      1         BB100201
originators:
globalId                     blocked primaryIp                       routes viaIp                    viaDev metric lastDesc lastRef
mlc1000.0AE58311046412F248CD 0       fd66:66:66:0:a2cd:efff:fe10:1   1      fe80::a2cd:efff:fe10:1   eth1   999M   3193     3
mlc1001.7A7422752001EC4AC4C8 0       fd66:66:66:0:a2cd:efff:fe10:101 0      ::                       ---    128G   3197     0
mlc1002.91DCF042934B5913BB00 0       fd66:66:66:0:a2cd:efff:fe10:201 1      fe80::a2cd:efff:fe10:201 eth1   999M   3196     3
mlc1003.09E796BC491D386248C3 0       fd66:66:66:0:a2cd:efff:fe10:301 1      fe80::a2cd:efff:fe10:201 eth1   576M   22       3

Only if relevant information exists for a requested type is available will it be shown. In this example no tunnels are configured locally nor are any tunnels offered by other nodes, so no tunnel information is shown.

The loop argument can be prepended to the connect argument to continuously show the requested information. Many of the long arguments are available via a short notation, like l for loop, c for connect, s for show, d for debug. And there is another shortcut summarizing my current favorite information types via debug level 8 The following commands do the same as above: bmx6 -lc status interfaces links originators tunnels or simply bmx6 -lcd8.

Description of the provided info:

interfaces: Followed by one line per configured interface
    dev: Interface name
    state and type: Whether the interface is UP or DOWN and its assumed link-layer type.
    rateMin and rateMax: Min- and maximum transmit rates assumed for this interface.
    llocalIp: IPv6 link-local address (used as source address for all outgoing protocol data).
    globalIp: Autoconfigured address used for sending network traffic via this interface and which is propagated to other nodes.
    multicastIp: Multicast IP (used as destination address for all bmx6 protocol traffic send via this interface).
    primary: Indicates whether the global ip of this interface is used as primary ip for this daemon.
links: Followed by one line per detected neighboring bmx6 node.
    globalId: GlobalId of that neighbor (see: Wiki ).
    llocalIp: Link-local IP of the neighbor's interface building the other side of the link.
    viaDev: Interface of this node for the link.
    rxRate: Measured receive rate in percent for the link.
    txRate: Measured transmit rate in percent for the link.
    bestTxLink: Indicates whether this link is the best link to a neighboring nodes.
    routes: Indicates for how much routes to other nodes this link is used.
    wantsOgms: Indicates whether the neighboring node has requested (this node) to propagate originator messsages (OGMs) via this link.
    nbLocalId: Neighbors local ID.
originators: Followed by one line per aware originator in the network (including itself).
    globalId: Global Id of that node (see: Wiki ).
    blocked: Indicates whether this node is currently blocked (see: Wiki ).
    primaryIp: The primary IP of that node.
    routes: Number of potential routes towards this node.
    viaIp: Next hops link-local IP of the best route towards this node.
    viaDev: Outgoing interface of the best route towards this node.
    metric: The end to end path metric to this node
    lastDesc: Seconds since the last description update was received (see: Widi )
    lastRef: Seconds since this node was referenced by any neighboring node (like last sign of life)

Quick summary of provided info:

  • Node mlc1001 uses one wired interface (eth1) which is up and actively used for meshing.
  • Node mlc1001 is aware of 2 neighbors and 4 nodes (originators) including itself.
  • The link quality metrics (rx and tx rate) to its neighbors are perfect (100%) and actively used (bestTxLink)
  • Routes to nodes mlc1000 and mlc1002 are via interface eth1, and directly to the neighbor's link-local address with a metric of 999M (nearly maximum tx/rx rate of the configured interface)
  • The route to node mlc1003 also uses interface eth1, but via the intermediate link-local address of neighbor mlc1002 (at least two hops to the destination node).

The following network topology can be deduced from this information (further links may exist):

mlc1000  mlc1001  mlc1002 - - - mlc1003
   |        |        |
   +--------+--------+

Simple Network Test

The above topology can be verified using traceroute6 specifying the primary IP address of the other nodes.

A ping to mlc1000's primary IP fd66:66:66:0:a2cd:efff:fe10:1 shows one hop:

root@mlc1001:~# traceroute6 -n -q 1 fd66:66:66:0:a2cd:efff:fe10:1
traceroute to fd66:66:66:0:a2cd:efff:fe10:1 (fd66:66:66:0:a2cd:efff:fe10:1), 30 hops max, 80 byte packets
 1  fd66:66:66:0:a2cd:efff:fe10:1  0.324 ms

To mlc1002's primary IP fd66:66:66:0:a2cd:efff:fe10:201 shows one hop:

root@mlc1001:~# traceroute6 -n -q 1 fd66:66:66:0:a2cd:efff:fe10:201
traceroute to fd66:66:66:0:a2cd:efff:fe10:201 (fd66:66:66:0:a2cd:efff:fe10:201), 30 hops max, 80 byte packets
 1  fd66:66:66:0:a2cd:efff:fe10:201  0.302 ms

To mlc1003's primary IP fd66:66:66:0:a2cd:efff:fe10:301 shows two hops:

root@mlc1001:~# traceroute6 -n -q 1 fd66:66:66:0:a2cd:efff:fe10:301
traceroute to fd66:66:66:0:a2cd:efff:fe10:301 (fd66:66:66:0:a2cd:efff:fe10:301), 30 hops max, 80 byte packets
 1  fd66:66:66:0:a2cd:efff:fe10:201  0.313 ms
 2  fd66:66:66:0:a2cd:efff:fe10:301  0.429 ms

Dynamic Reconfiguration

Most bmx6 parameters can be applied not only at startup, but also dynamically to an already running main daemon, using the --connect command. For example interfaces can be added, removed, or specified with more details: The following example removes interface eth1 and adds eth2 with a max rate of 100 Mbits (overwriting the default assumption of 1000Mbits for ethernet interfaces).

bmx6 -c dev=-eth1 dev=eth2 /rateMax=100000
bmx6 -cd8

Checking new status of interfaces, links, and originator:

root@mlc1001:~# bmx6 -cd8
status:
version        compatibility codeVersion globalId                     primaryIp                       myLocalId uptime     cpu nodes
BMX6-0.1-alpha 16            9           mlc1001.7A7422752001EC4AC4C8 fd66:66:66:0:a2cd:efff:fe10:102 06100101  0:02:26:00 0.1 4
interfaces:
devName state type     rateMin rateMax llocalIp                    globalIp                           multicastIp primary
eth2    UP    ethernet 100M    100M    fe80::a2cd:efff:fe10:102/64 fd66:66:66:0:a2cd:efff:fe10:102/64 ff02::2     1
links:
globalId                     llocalIp               viaDev rxRate txRate bestTxLink routes wantsOgms nbLocalId
mlc1000.0AE58311046412F248CD fe80::a2cd:efff:fe10:2 eth2   89     88     1          3      1         9B100001
originators:
globalId                     blocked primaryIp                       routes viaIp                  viaDev metric lastDesc lastRef
mlc1000.0AE58311046412F248CD 0       fd66:66:66:0:a2cd:efff:fe10:1   1      fe80::a2cd:efff:fe10:2 eth2   81757K 18       0
mlc1001.7A7422752001EC4AC4C8 0       fd66:66:66:0:a2cd:efff:fe10:102 0      ::                     ---    128G   80       0
mlc1002.91DCF042934B5913BB00 0       fd66:66:66:0:a2cd:efff:fe10:201 1      fe80::a2cd:efff:fe10:2 eth2   83620K 14       4
mlc1003.09E796BC491D386248C3 0       fd66:66:66:0:a2cd:efff:fe10:301 1      fe80::a2cd:efff:fe10:2 eth2   81488K 9        0

It can be seen that:

  • Interface eth1 has been replaced by eth2 with a lower rate.
  • The primary IP of the node has changed (using the autoconfigured IP from eth2.
  • The old links (via eth1) are removed and a single new link via eth2 to mlc1000 has been detected
  • All routes are now going via eth2 and mlc1000's link-local IP fe80::a2cd:efff:fe10:2

Concepts

Global ID

Each bmx6 node creates during its initialization (booting) a global ID for itself. This ID is created as a concatenation of the node's hostname and a random value. In the above given example with node hostname: "mlc1001" the globalID is: mlc1001.7A7422752001EC4AC4C8 When the bmx6 daemon restarts the hostname will remain. But the rand part will change. As a consequence, the restarted node will appear as a new node to other nodes in the mesh while the old Global ID is still present in their node table. Since both node IDs are announcing the same resources (eg the same primary IP), the ID that appears later will be blocked until the state maintained for the first ID expires.

Descriptions

Instead of propagating individual routing updates for each announced network and interface address, each bmx6 daemon summarizes this and other node specific attributes into a single node-specific description. A specific description is propagated only once to all other nodes. Subsequent routing updates are referencing to the corresponding description with it's hash. If a node is reconfigured, for example because its interfaces change or a new network shall be announced, than also the node's description changes. Other nodes are becoming aware of the changed attributes of a reconfigured node by receiving a corresponding description update. Subsequent references to this node will use the hash of the new description.

Because the description is designed very generic it can be easily used to piggyback other non-routing specific data. For example the bmx6-sms plugin is taking advantage of this option by adding arbitrary short messages data to the node's description.

Currently there is a limit for the total size of a description of 1400 bytes. While this is more than sufficient for quite a number of interfaces and announced networks per node, it is critical few when considering a gateway node with BGP route exchange that is announcing 100eds of networks.

Blocked Nodes

Nodes may be blocked by other nodes. When a node is blocked no routing updates (OGMs) of the blocked node are propagated by the blocking node. The decision for blocking another node is done locally based on the detection of more than one node announcing the same unique resource. This happens if two nodes are declaring themselves as the owner of a unique resource. Then one of those two nodes (usually the latter) is blocked to avoid the propagation of conflicting allocations (and ambiguous forwarding state). Duplicate address usage is the most common reason for such events which happens if two nodes are using (and announcing) the same primary IPs. Another typical scenario causing such case temporary is the rebooting of a node. Once a bmx6 daemon restarts it appears as a new node (with a new random part of it's global ID) to the network but (due to a typically persistant configuration) announcing the same address as the previous process. Since the resources allocated by the previous resources are still in the database of other nodes in the mesh they will block the new process until this information expires (by default after 100 seconds).

Address auto and manual configuration

By default bmx6 autoconfigures all configred interface by combining a default prefix (fd66:66:66::/64) with the EUI64 suffix (the suffix creation is currently reconsidered and may change soon). The same first 56 bits but extended with 0xff00 are also used to create tunnel interfaces.

There are different options to controll the auto configuration.

  1. A different auto-configuration prefix can be used using the

     --ipAutoPrefix 
    option given with a /56 prefix.

  2. Auto configuratin can be disabled using the

     --globalPrefix 
    option. Then bmx6 checks if an ip in this range is alredy configured on the interfaces and uses it. If no IP is configured in the given range then the inteface will NOT be used.

Unicast Host Network Announcements (UHNA)

A Host Network Announcements (HNA) describes the advertisement of IP addresses and networks by a node to other nodes in the mesh. Typically (but not with BMX6), several nodes can announce the same or overlapping HNAs at the same time. Announced networks do overlap if they are equal or one being a subset of another (eg. 10.1.1.0/24 is a subset and overlapped by 10.1.0.0/16). Packets with a destination address matching an announced networks will be routed toward any node that originated a corresponding HNA. Therefore these HNA types may also be called anycast HNA.

In bmx6, HNAs have an unicast nature (UHNAs) because each network can only be announced once and announced networks MUST NOT overlap (See also Wiki). This way it can be ensured that the destination of an UHNA routed packet is exactly known.

In a sense the origination and propagation (by intermediate nodes) of UHNA announcements can be thought of a promise that guarantees:

  1. All packets with a destination address matching an announced UHNA network will be routed exactly to the node (with the global ID) that originated the UHNA and
  2. each node on the forwarding path towards the originator of the UHNA is supporting this promise.

By default, Bmx6 only announces primary and non-primary interface addresses via UHNAs. The auto address configuration ensures that interface addresses are unique.

Using UHNAs for the announcements of networks requires a strict coordination to ensure that no network is announced twice.

Technically, multiple UHNAs, each wrapped into a single message, are aggregated into a UHNA frame and attached to the description of a node.

If Bmx6 is configured in IPv6 mode only IPv6 UHNAs can be announced and in IPv4 mode only IPv4 UHNAs UHNA Configuration

The announcement of UHNAs can be configured with the --unicastHna or -u parameter followed by a network specification in ip/prefixlen notation. By default all interface addresses are announced via UHNAs. However, this can be disabled by setting the --dev subparameter /announce or /a to 0.

The following example reconfigures an already running bmx6 daemon (in IPv6 mode) to UHNA announce the network fd00:ffff:ffff:ffff::/64 and fd01:ffff:ffff::/48. By omitting the --connect / -c parameter, the same could be configured as startup parameter for bmx6.

bmx6 -c u=fd00:ffff:ffff:ffff::/64 u=fd01:ffff:ffff::/48

An already active announcement can be removed by preceeding the network with the - char:

bmx6 -c u=-fd00:ffff:ffff:ffff::/64

Before bmx6 accepts a dynamically configured UHNA announcement it checks if this UHNA is not overlapping with an already existing UHNA announcement form another node. If this is the case the configuration will fail. To check if a chain of dynamic commands would be accepted by a bmx6 daemon without actually applying it, the --test command may follow the --connect command.

Tunnel Announcements

Tunnel announcements offer an alternative mechanism to propagate routes. Tunnel announcements are currently only implemented for Bmx6-IPv6 mode. However, in IPv6 mode IPv6 and IPv4 networks can be announced. In contrast to UHNAs, using tunnel announcements, the same or overlapping networks can be announced from different nodes. Tunnel announcements are an offer from the originating node to other nodes. Other nodes can take the offer or not. For example several nodes in a network may offer to share their DSL connection by doing a default-route (0.0.0.0/0 or ::/0) tunnel announcement. Other nodes looking for a route to the internet (a default route) can choose between the multiple offers by establishing a tunnel to one specific of the offering nodes. Therefore an unidirectional (onw-way) tunnel is established from the searching to the offering node. At the searching node, the remote (outer) tunnel address is configured with an UHNA address (usually the primary address) of the offering node. The networks advertised with the tunnel announcements are configured at the client side as routes via (into) the unidirectional tunnel.

This way, each node can make an individual choice between networks offered via tunnel announcements. The automatic selection can be specified via a policy description that considers parameters such as advertised bandwidth, path metric, trust in specific GW nodes, hysteresis, ... . Since an UHNA address is used as the outer (remote) tunnel address, the client end of the tunnel can be sure that all packets routed into the tunnel will indeed end up at the intended GW node (see Wiki).

Technically, multiple tunnel announcements, each wrapped into a single tun4/6in6-net message, are aggregated into a tun4/6in6-net frame and attached to the description of a node.

Tunnel announcements are also used for redistributing routes from other routing protocols (see Wiki) into a bmx6 zone. Therefore, each announcements message is decorated with a route-type field indicating the routing protocol that exported the route for being redistributed.

Tunnel requirements

The following Linux-kernel modules are needed for tunnel-based overlay networking:

  • ipv6
  • tunnel6
  • ip6_tunnel

Tunnel Configuration and Debugging

In general, a specific tunnel configuration is described from two perspectives:

  • Gateway (GW) nodes or just GWs are offering GW services to networks via the advertizement of tunnel announcements and the provisioning of tunnel-end-points.

  • GW-client nodes (or just GW-clients) that are searching for GWs with tunnel endpoints and routing services to networks.

A node can (and usually is) operating in both modes (as GW and as GW-client). But regarding a specific network each node is operating either in GW mode (thus, offering GW-services to that network) or in GW-client mode (thus, searching and using GW-services to that network)!

A quick and simple tunnel configuration example is given here. Further details and options are described in the next Sections. The full set of available options for configuring tunnels is given via the build-in --help and --verboseHelp commands

  • First make your own tunnel addresses known and reachable for other nodes, eg: bmx6 -c tunDev=Default /tun4Address=10.254.10.123/32 /tun6Address=2012:1234:5678:123::1/64

  • Second, configure the automatic establishment of outgoing tunnels to other nodes by searching and selecting any kind of announcement: bmx6 -c tunOut=v4Default /network=0.0.0.0/0 tunOut=v6Default /network=::/0

  • Optionally, check the currently offered tunnel announcements of other GW nodes and the selected tunnel routes by this node with: bmx6 -c show=tunnels

Remark: Since master commit f2fd75072f7dc4738069be6c69625419b9cc7767 the syntax for configuring tunnels has changed. In the following the new syntax is explained. For the old syntax please use the build-in --help and --verboseHelp of the binary you are using

Tunnel Device Configuration

Operation in GW and/or GW-client mode implies the configuration of a bmx6 tunnel device and the IPv4 and/or IPv6 addresses that shall be used for tunnel traffic. The selection of these addresses should be coordinated with:

  • the mesh community because conflicting tunnel address usage will cause problems for the conflicting nodes
  • GW administrators because (depending on the GW connection to other networks) only specific addresses may be routable and considered to be originated from the bmx6 cloud.

The command

bmx6 -c tunDev=Default /tun4Address=10.254.10.123/32 /tun6Address=2012:1234:5678:123::1/64

dynamically

  • configures a linux ip4/6in6 tunnel device called bmx6Default (check it with command: ip link show).
  • assignes the address 10.254.10.123 and 2012:1234:5678:123::1 to the tunnel interface and uses them for outgoing tunnel traffic.
  • enables GW-mode for the specified networks: Makes a tunnel announcement so that other nodes can select it for tunneling packets to this node.

Now other nodes can send tunneled packets to this node via the unidirectional tunnel end point offered by this node.

But for bidirectional tunnel communication with any another node also a backwards tunnel is needed (an unidirectional tunnel from this node to the other node).

The automatic selection and establishemt of tunnels to other nodes is achieved with the GW-client mode as described in more derail in the next Section.

Gateway-Client Nodes

The configuration of GW clients can be simple but also, depending on the preferences for a GW-selection policy, very complex. Through the configuration of the mandatory tunDev and it's addresses (see above), each GW client node is also a GW node to its own (usually small) tunnel address space.

In the following simple example a GW-client node is searching for any other kind of offered IPv4 and v6 tunnels:

bmx6 -c tunOut=v4Default /network=0.0.0.0/0 tunOut=v6Default /network=::/0

With the above configured tunnel selection policy, tunnels are selected in the following order:

  1. prefix-length of announced tunnels (networks that are more specific than others).
  2. the resulting tunnelMetric (combination of the advertised bandwidth, path metric in the bmx6 cloud, and locally specified prefereces like hysteresis or bonus)

The disadvantage of this simple config is that other nodes can easily redirect your tunnel selections to specific networks by announcing more precise tunnel networks (larger prefix length). To prevent this, the selection policy can be split into several and more precise search directives.

Imagine the following address assignment policy for IPv4 tunnel addresses in a mesh cloud (the general idea can be straight translated to IPv6).

  • Nodes in the mesh cloud announce their private and local address ranges with a prefix length of 24 and somewhere in the range of 10.254.0.0/16.

  • Announcements of this type should always be preferred, even if any of the following announced types has a better end-to-end metric or more precise announcement.

    bmx6 -c tunOut=v4Nodes /network=10.254.0.0/16 /minPrefixLen=24 /maxPrefixLen=24 /ipmetric=2001
    
  • Some BGP GW nodes are connected to other mesh clouds/areas of the same overall community network. These clouds are operating in a different IPv4 range (than 10.254.0.0/16) but always somewhere in the range of 10.0.0.0/8. Route announcements of this type should be preferred over the announcement of a default route.

    bmx6 -c tunOut=v4Clouds /network=10.0.0.0/8 /maxPrefixLen=16 /bgp=1
    
  • Some DSL GW nodes are offering to share their DSL line and are announcing a default route (0.0.0.0/0).

  • Further, to mitigate the effects of GW switching between GWs having a similar end-to-end metric a GW switch should only happen if the other GW is at least 30% better.

    bmx6 -c tunOut=v4Default /network=0.0.0.0/0 /maxPrefixLen=0 /hysteresis=30 # refine the above configured v4 tunnel search
    
  • In case my node is directly connected to a DSL gateway and gets a automatically (dhcp) configured default route in the main routing table (use: ip route show table main ). then this route should be preferred and should NOT clash with default tunnel routes configured by bmx6.

  • Therefore move all bmx6 tunnel routes to 0.0.0.0/0 into a separate routing table with lower lookup prioriy (check with: ip rule show; ip route show table 150)

    bmx6 -c tunOut=v4Default /network=0.0.0.0/0 /maxPrefixLen=0 /hysteresis=30 /tableRule=50000/150 # again refine the above default search
    
  • The default route announcements from two well known GWs (with hostname pepe and paula) should be strictly preferred over unknown GWs.

  • So, if available, move them to new table (with lower priority than main and higher priority than used for the backup tunnel rule configured above)

    bmx6 -c tunOut=v4DefaultPepe  /network=0.0.0.0/0 /maxPrefixLen=0 /gwName=pepe  /hysteresis=30 /tableRule=40000/140
    bmx6 -c tunOut=v4DefaultPaula /network=0.0.0.0/0 /maxPrefixLen=0 /gwName=paula /hysteresis=30 /tableRule=40000/140
    
  • Finally, GW Paula turned out to be more stable. Therefore I want to prefer GW Paula over Pepe:

    bmx6 -c tunOut=v4DefaultPaula /network=0.0.0.0/0 /maxPrefixLen=0 /gwName=paula /hysteresis=30 /bonus=100
    

Gateway Nodes

The advertisement of a tunnel endpoint to a network can be configured with the --tunIn= and /network= argument and an optional bandwidth specification (given as bits per second) using the /bandwidth or /b sub parameter. Announcement can be removed by preceeding the name argument with a '-' char. The following command dynamically configures the advertisement of the following routes:

  • An IPv4 default route 0.0.0.0/0 with a bandwidth of 32 Mbps.
  • A more specific route to 10.10.0.0/16 with a bandwidth of 10 Mbps (eg: a local v4 Network).
  • An IPv6 route to the [RFC 4291] designated 2000::/3 global unicast address space with a bandwidth of 16 Mbps.
  • A more specific route to the 2012:1234::/32 IPv6 space at 10 Mbps (eg: a local v6 Network).
    bmx6 -c tunIn=def4Offer /n=0.0.0.0/0 /b=32000000  tunIn=local4 /n=10.10.0.0/16 /b=10000000  tunIn=def6Offer /n=2000::/3 /b=16000000  tunIn=local6 /n=2012:1234::/32 /b=10000000
    

Tunnel Status Information

Tunnel status information can be accessed with the --tunnels or --show=tunnels parameters.

Bmx6 Plugins

Compile and Install

To compile and install bmx6 daemon and all bmx6 plugins simply do:

make build_all
sudo make install_all

However. specific requirements may need to be fulfilled for some plugins in order to compile correctly. These requirements are described in the corresponding plugin section.

Config Plugin

Requirements

uci libs are needed for the bmx6-config plugin. To install it do:

wget http://downloads.openwrt.org/sources/uci-0.7.5.tar.gz
tar xzvf uci-0.7.5.tar.gz
cd uci-0.7.5
make
sudo make install

Depending on your system there happens to be an error during compilation. Then edit cli.c and change line 465 to: char *argv[MAX_ARGS+2];

Compile and Install

make -C lib/bmx6_uci_config/
sudo make -C lib/bmx6_uci_config/ install

Usage

Json Plugin

Requirements

json-c for bmx6_json plugin (debian package: libjson0 libjson0-dev)

json-c developer libs are needed! For further reading check: http://json.org/ or https://github.com/jehiah/json-c

Note for debian sid: The debian package libjson0-dev 0.10-1 seems to miss the file /usr/include/json/json_object_iterator.h Manually copying it from the below mentioned json-c_0.10.orig.tar.gz archive helps.

To install manually (only if NOT installed via debian or other package management system):

wget http://ftp.de.debian.org/debian/pool/main/j/json-c/json-c_0.10.orig.tar.gz
tar xzvf json-c_0.10.orig.tar.gz
cd json-c..
./configure ; make ; make install; ldconfig

Compile and Install

To compile and install only the bmx6 json plugins:

make -C lib/bmx6_json/
sudo make -C lib/bmx6_json/ install

Usage

SMS Plugin

This plug-in uses routing packets to transmit any information from one node to the whole network. The good point is that propagation works even if there is no continuous data- path. Even though the WiFi network is under bad conditions (because the Wireless noise, distance between nodes, etc...), the data will be propagated. However in the current implemen- tation, there exist a maximum size limit of 240 Bytes for each file.

The API of the sms plug-in is very simple. It simply clones the content of one or more files given by one node to all other nodes. All other nodes can do the same. Once started, each node will have two directories:/var/run/bmx6/sms/rcvdSms and /var/run/bmx6/sms/sendSms. Files put into the sendSms folder will be cloned to all other nodes inside rcvdSms folder. QMP is using this feature for several things. The positioning Map information is transmitted using it. There is a chat in web interface which uses it too. And in the future we are planning to use it for more purposes like statistics, captive portal, MAC filter rules, etc...

Table plugin

This plug-in can be used to automatically announce routes from specific routing tables. For example to dynamically announce (redistribute) routes from another routing protocol.

Usage

To use the bmx6 table plugin it must be loaded during bmx6 daemon startup with the plugin=bmx6_table.so argument. Alternatively a plugin section can be defined in the bmx6 config file like this:

config 'plugin'
        option 'plugin' 'bmx6_table.so'

Once the plugin is successfully loaded, the new parameters for redistributing routes from specific tables are enabled.

A full documentation of the table-related parameters is available via the --help and --verboseHelp /r=1 option.

Configuring route redistribution

Redistribution of routes is configurable with the --redistTable parameter. Similar to the --tunIn parameter, --redistTable must be given with an arbitrary name for referencing to a specific redistribution directive and further sub-criterias.

Mandatary sub-criterias are /table and at least one route of the available types ( /kernel, /boot, /static ) Typical further but optional sub parameters are: /network, /minPrefixLen, /maxPrefixLen, /aggregatePrefixLen, /bandwidth The following example automatically and dynamically announces (and removes announcements of) routes as they appear/disappear in routing table 100 and that match the criterias of:

  • being a sub network of 192.168.0.0/16
  • have a prefix length >= 24
  • are configured as type kernel or boot (manually configured routes via the ip command will appear by default as type boot, eg: ip r add 192.168.254.2/31 via 10.0.0.1 table 100

If routes matching these criterias exist, then:

  • They are announced with a bandwidth of 1Mbit
  • Subsequent routes are aggregated down to a minimum prefix length of 24
    bmx6 -c \
    redistTable            otherProtocol        \
      /network            192.168.0.0/16       \
      /table              100                  \
      /aggregatePrefixLen 24                   \
      /minPrefixLen       24                   \
      /kernel             1                    \
      /boot               1                    \
      /bandwidth          1000000
    

Quagga Plugin

The bmx6 quagga plugin can be used to exchange routes with a quagga/zebra daemon. Both, export and redistribution of routes is supported.

Requirements, Compile, and Install

Quagga

Quagga version 0.99.21 must be patched for bmx6 support.

The bmx6 directory lib/bmx6_quagga/patches/ contains patches to enable quagga for bmx6 support. The following example provides instructions for obtaining, patching, compiling, and installing quagga:

wget http://download.savannah.gnu.org/releases/quagga/quagga-0.99.21.tar.gz
tar xzvf quagga-0.99.21.tar.gz
cd quagga-0.99.21
patch -p1 < ../bmx6/lib/bmx6_quagga/patches/quagga-0.99.21.tar.diff
./configure
make
sudo make install

For further instructions to obtain, patch, compile, and install quagga please have a look at: the file lib/bmx6_quagga/patches/README in the bmx6 sources.

Bmx6

To compile and install the bmx6 part of the quagga plugin simply do:

make -C lib/bmx6_quagga/
sudo make -C lib/bmx6_quagga/ install

Usage

To use the bmx6 quagga plugin it must be loaded during bmx6 daemon startup with the plugin=bmx6_quagga.so argument. Alternatively a plugin section can be defined in the bmx6 config file like this:

config 'plugin'
        option 'plugin' 'bmx6_quagga.so'

Once the plugin is successfully loaded, the bmx6 daemon will try to connect with the zebra process (via the ZAPI socket) and new parameters for exchanging routes with quagga/zebra daemon are enabled.

A quick documentation of the quagga-related parameters is available via the --help and --verboseHelp option. If the quagga-enabled daemon is already running bmc6 -c verboseHelp /r=1 will print all currently supported parameters. Redistributing routes (from quagga/zebra to bmx6)¶

Redistribution of routes is configurable with the --redistribute parameter. Similar to the --tunIn parameter, --redistribute must be given with an arbitrary name for referencing to a specific redistribution directive and further sub-criterias.

Further mandatory sub-parameters are /bandwidth and at least one (to-be redistributed route type). The following route types exist:

  /system                           def: 0       range: [ 0 , 1 ]
  /kernel                           def: 0       range: [ 0 , 1 ]
  /connect                          def: 0       range: [ 0 , 1 ]
  /rip                              def: 0       range: [ 0 , 1 ]
  /ripng                            def: 0       range: [ 0 , 1 ]
  /ospf                             def: 0       range: [ 0 , 1 ]
  /ospf6                            def: 0       range: [ 0 , 1 ]
  /isis                             def: 0       range: [ 0 , 1 ]
  /bgp                              def: 0       range: [ 0 , 1 ]
  /babel                            def: 0       range: [ 0 , 1 ]
  /hsls                             def: 0       range: [ 0 , 1 ]
  /olsr                             def: 0       range: [ 0 , 1 ]
  /batman                           def: 0       range: [ 0 , 1 ]

Only quagga/zebra routes types that are explicitly specified will be redistributed to the bmx6 network. In addition, one usually wants to filter out networks from being redistributed based on their prefix. Therefore the sub parameters /network, /minPrefixLen, and /maxPrefixLen can be used in the same way as for the --tunOut parameter.

Route Aggregation

By default, maximum aggregation of to-be redistributed routes is enabled. This means that to-be redistributed neighboring and overlapping networks with the same route type and bandwidth are aggregated if possible. The extend of aggregation can be controlled with the /aggregatePrefixLen sub-parameter. The given value limits the aggregation to a minimum prefix length. The default of 0 defines maximum aggregation whenever possible which may not be wanted.

For example the GW node may be configured to redistribute the following routes:

    10.254.20.1/32
    10.254.20.0/24
    10.254.21.0/24
    10.254.22.0/24
    0.0.0.0/0

The following bmx6 configuration would aggregate all 5 routes into a single 0.0.0.0/0 tunnel announcement since 0.0.0.0/0 is overlapping any other more-specific route:

redistribute=ipv4 /bandwidth=10000000 /kernel=1 /aggregatePrefixLen=0

This aggregation may be too generic since GW-client nodes are usually looking for more specific routes to specific destination. The following configuration would aggregate only routes with a prefix-len larger than 16:

redistribute=ipv4 /bandwidth=10000000 /kernel=1 /aggregatePrefixLen=16

Resulting in the following aggregations:

  • 10.254.20.1/32: Aggregated (sub-network of 10.254.20.0/24)! NOT announced!
  • 10.254.20.0/24: Aggregated with 10.254.21.0/24! Announced as 10.254.20.0/23
  • 10.254.21.0/24: Aggregated with 10.254.20.0/24! Announced as 10.254.20.0/23
  • 10.254.22.0/24: Not aggregatable into larger network! Announced as is!
  • 0.0.0.0/0: Not aggregated (prefix-len smaller than /aggregatePrefixLen=16)! Announced as is!

Exporting routes (from bmx6 to quagga/zebra)

For exporting routes received as bmx6 tunnel announcements, the /exportDistance can be used as a subparameter of the --tunOut parameter. The default value of /exportDistance is 256 which is considered as infinit or disabled. Any lower configured value will export the corresponding outgoing tunnel (once it becomes active) with the given distance to quagga/zebra.

A GW node usually only wants to export bmx6 routes that were announced by other (non-GW) bmx6 nodes in the mesh.

In the following example there are 3 other bmx6 nodes, each tunnel announcing a private /32 network.

The given parametrization configures a GW node to search, establish related tunnels, and export all tunnel announcements for other bmx6 daemons that have a prefix-length smaller that /27 and fall into the network range of 10.254.0.0/16:

plugin=bmx6_quagga.so tunOut=privV4Nets /network=10.254.0.0/16 /minPrefixLen=27 /exportDistance=0

Checking the export from the quagga perspective show the following:

root@mlc1001:~# telnet localhost zebra
Trying ::1...
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.

Hello, this is Quagga (version 0.99.21).
Copyright 1996-2005 Kunihiro Ishiguro, et al.

User Access Verification
Password:

Router> show ip route
Codes: K - kernel route, C - connected, S - static, R - RIP,
       O - OSPF, I - IS-IS, B - BGP, H - HSLS, o - OLSR,
       b - BATMAN, x - BMX6, A - Babel,
       > - selected route, * - FIB route

K>* 0.0.0.0/0 via 10.0.0.1, eth0
C>* 10.0.0.0/11 is directly connected, eth0
x>* 10.254.10.0/32 [0/1024] is directly connected, bmx6_out0000, 00:03:24
C * 10.254.10.1/32 is directly connected, bmx6_out0003
C * 10.254.10.1/32 is directly connected, bmx6_out0002
C * 10.254.10.1/32 is directly connected, bmx6_out0001
C * 10.254.10.1/32 is directly connected, bmx6_out0000
C>* 10.254.10.1/32 is directly connected, bmx6_in0000
x>* 10.254.10.2/32 [0/1024] is directly connected, bmx6_out0001, 00:03:24
x>* 10.254.10.3/32 [0/1024] is directly connected, bmx6_out0002, 00:03:24
x>* 10.254.10.4/32 [0/1024] is directly connected, bmx6_out0003, 00:03:24
C>* 127.0.0.0/8 is directly connected, lo