BIRD <author> Pavel Machek <tt/pavel@ucw.cz/ <date>2000 <abstract> This document contains documentation for BIRD Internet Routing Daemon </abstract>  <toc>  <sect>Introduction <sect1>What is BIRD <label id="intro"> You may wonder what 'BIRD' means. It is acronym of 'BIRD Internet Routing Daemon', and we think that's cool name. Routers are devices for connecting networks. Router has several network interfaces, listens for packets on them, and forwards packets closer to their destination. For simple situations, such as tree, static routing is enough and your kernel can do routing for you. For more complicated topologies, you need routers to talk to eachother (in order to discover alternate ways to deliver packets after network failure). Routing deamon is the beast that accomplishes this communication. It does not forward packets itself, but it tells kernel where to forward them. Its task is similar to what firmware of Cisco routers does, or what gated <HTMLURL URL="http://www.gated.org/"> or GNU zebra <HTMLURL URL="http://www.zebra.org/"> does. However, you can not run Cisco's firmware on "normal" computer and gated is really hard to configure and comes under wrong license. BIRD is being developed on Charles University, Prague, and can be freely distributed under terms of GNU General Public License. BIRD is designed to be portable; it is primarily developed on Linux. <sect1>About this documentation This documentation can have 4 forms: sgml (this is master copy), html, ASCII text (generated from html) and dvi/postscript (generated from sgml using sgmltools). You should always edit master copy, it is slightly modified linuxdoc dtd. Anything in <descrip> tags is considered definition of configuration primitives, <cf> is fragment of configuration within normal text, <m> is "meta" information within fragment of configuration -- something in config which is not keyword. <sect1>About routing tables Bird has one or more routing tables. Each routing table contains list of known routes. Each route has certain attributes, most important is prefix of network this route is for. Routing table maintains more than one entry for network, but at most one entry for one network and one protocol. The entry with biggest preference is used for routing. If there are more entries with same preference and they are from same protocol, protocol decides (typically according to metrics). You can get list of route attributes in "Route attributes" section in filters. <sect>Configuration <sect1>Introduction BIRD is configured using text configuration file. At startup, BIRD reads <file/bird.conf/ (unless -c command line parameter is given). Configuration may be changed on user request: if you modify config file and then signal BIRD with SIGHUP, it will adjust to new config. There's BIRD client, which allows you to talk with BIRD in more extensive way than just telling it to reconfigure. BIRD writes messages about its work to log files or syslog (according to config). Bird is configured using text configuration file. At startup, bird reads <file/bird.conf/ (unless -c command line parameter is given). Everything on a line after <cf/#/ is a comment, whitespace is ignored, C-style comments <cf>/* comment */</cf> are also recognized. If there's variable number of options, it is grouped using <cf/{ }/ brackets. Each option is terminated by <cf/;/. Really simple configuration file might look like this: <code> protocol kernel { persist; # Don't remove routes on BIRD shutdown scan time 20; # Scan kernel routing table every 20 seconds export all; # Default is export none } protocol device { scan time 10; # Scan interfaces every 10 seconds } protocol rip { export all; import all; } </code> <sect1>Global options <descrip> <tag>log "<m/filename/"|syslog|stderr all|{ <m/list of classes/ }</tag> set logging of classes (either all or <cf/{ error, trace }/ etc.) into selected destination. Classes are: <cf/info/, <cf/warning/, <cf/error/, <cf/fatal/ for messages about local problems <cf/debug/ for debugging messages, <cf/trace/ when you want to know what happens on network, <cf/remote/ for messages about misbehavior of remote side, <cf/auth/ about authentication failures, <cf/bug/ for internal bugs of BIRD. You may specify more than one <cf/log/ line to log to multiple destinations. <tag>debug protocols all|off|{ states, routes, filters, interfaces, events, packets }</tag> sets global default of protocol debugging options. <tag>filter <m/name/{ <m/commands/ }</tag> define filter. You can learn more about filters in next chapter. <tag>protocol rip|ospf|bgp|... <m/[name]/ { <m>protocol options</m> }</tag> define protocol instance, called name (or called something like rip5 if you omit name). You can learn more about configuring protocols in their own chapters. You can run more than one instance of most protocols (like rip or bgp). <tag>define constant = expression</tag> define constant. You can use it later in every place you could use simple integer. <tag>router id <m/IPv4 address/</tag> set router id. Router id needs to be world-wide unique. It is usually one of router's IPv4 addresses. <tag>table <m/name/</tag> create new routing table. <tag>eval <m/expr/</tag> evaluates given filter expression. It is used for testing. </descrip> <sect1>Protocol options Several options are per-protocol, but all protocols support them. They are described here. <descrip> <tag>preference <m/expr/</tag> sets preference of routes generated by this protocol. <tag>disabled</tag> disables given protocol. You can disable/enable protocol from command line interface without needing to touch config. <tag>debug <m/setting/</tag> this is similar to global debug setting, except that it only affects one protocol. Only messages in selected debugging categories will be written to logs. <tag>import <m/filter/</tag> filter can be either either <cf> { <m>filter commands</m> }</cf> or <cf>filter <m/name/</cf>. Import filter works in direction from protocol to main routing table. <tag>export <m/filter/</tag> This is similar to <cf>export</cf> keyword, except that it works in direction from main routing table to protocol. <tag>table <m/name/</tag> Connect this protocol to non-default table. </descrip> There are per-protocol options that give sense only with certain protocols. <descrip> <tag>passwords { password "<m/password/" from <m/time/ to <m/time/ passive <m/time/ id <m/num/ [...] }</tag> specifies passwords to be used with this protocol. Passive time is time from which password is not announced but is allowed. id is password id, as needed by certain protocols. <tag>interface "<m/mask/"|<m/prefix/ [ { <m/option/ ; [ ... ] } ]</tag> specifies, which interfaces this protocol is active at, and allows you to set options on interface-by-interface basis. Mask is specified in shell-like patters, thus <cf>interface "*" { mode broadcast; };</cf> will start given protocol on all interfaces, with <cf>mode broadcast;</cf> option. </descrip> <sect>Filters <sect1>Introduction BIRD contains rather simple programming language. (No, it can not yet read mail :-). There are two objects in this language: filters and functions. Filters are called by BIRD core when route is being passed between protocol and main routing table, and filters may call functions. Functions may call other functions, but recursion is not allowed. Filter language contains control structures such as if's and switches, but it allows no loops. Filters are interpreted. Filter using many features can be found in <file>filter/test.conf</file>. You can find sources of filters language in <file>filter/</file> directory. <file>filter/config.Y</file> contains filter grammar, and basically translates source from user into tree of <cf>f_inst</cf> structures. These trees are later interpreted using code in <file>filter/filter.c</file>. Filters internally work with values/variables in <tt>struct f_val</tt>, which contains type of value and value. Filter basically gets the route, looks at its attributes and modifies some of them if it wishes. At the end, it decides, whether to pass change route through (using <cf/accept/), or whether to <cf/reject/ given route. It looks like this: <code> filter not_too_far int var; { if defined( rip_metric ) then var = rip_metric; else { var = 1; rip_metric = 1; } if rip_metric > 10 then reject "RIP metric is too big"; else accept "ok"; } </code> As you can see, filter has a header, list of local variables, and body. Header consists of <cf/filter/ keyword, followed by (unique) name of filter. List of local variables consists of pairs <cf><M>type name</M>;</cf>, where each pair defines one local variable. Body consists of <cf> { <M>statements</M> }</cf>. Statements are terminated by <cf/;/. You can group several statements into one by <cf>{ <M>statements</M> }</cf> construction, that is useful if you want to make bigger block of code conditional. There are two special filters, <cf/all/ (which accepts all routes) and <cf/none/ (which rejects all routes). Bird supports functions, so that you don't have to repeat same blocks of code over and over. Functions can have zero or more parameters, and can have local variables. Function basically looks like this: <code> function name () int local_variable; { local_variable = 5; } function with_parameters (int parameter) { print parameter; } </code> Unlike C, variables are declared after function line but before first {. You can not declare variables in nested blocks. Functions are called like in C: <cf>name(); with_parameters(5);</cf>. Function may return value using <cf>return <m/[expr]/</cf> syntax. Returning value exits from current function (this is similar to C). Filters are declared in similar way to functions, except they can not have explicit parameters. They get route table entry as implicit parameter. Route table entry is passed implicitly to any functions being called. Filter must terminate with either accept or reject statement. If there's runtime error in filter, route is rejected. <sect1>Data types Each variable and each value has certain type. Unlike C, filters distinguish between integers and booleans (that is to prevent you from shooting in the foot). <descrip> <tag/bool/ this is boolean type, it can have only two values, <cf/TRUE/ and <cf/FALSE/. Boolean is not compatible with integer and is the only type you can use in if statements. <tag/int/ this is common integer, you can expect it to store signed values from -2000000000 to +2000000000. <tag/pair/ this is pair of two short integers. Each component can have values from 0 to 65535. Constant of this type is written as <cf/(1234,5678)/. <tag/string/ this is string of characters. There are no ways to modify strings in filters. You can pass them between functions, assign to variable of type string, print such variables, but you can not concatenate two strings (for example). String constants are written as <cf/"This is a string constant"/. <tag/ip/ this type can hold single ip address. Depending on version of BIRD you are using, it can be IPv4 or IPv6 address. IPv4 addresses are written (as you would expect) as <cf/1.2.3.4/. You can apply special operator <cf>.mask(<M>num</M>)</cf> on values of type ip. It masks out all but first <cf><M>num</M></cf> bits from ip address. So <cf/1.2.3.4.mask(8) = 1.0.0.0/ is true. <tag/prefix/ this type can hold ip address, prefix len pair. Prefixes are written as <cf><M>ipaddress</M>/<M>pxlen</M></cf>, or <cf><m>ipaddress</m>/<m>netmask</m></cf> There are two special operators on prefix: <cf/.ip/, which separates ip address from the pair, and <cf/.len/, which separates prefix len from the pair. <tag/int|ip|prefix|pair set/ filters know four types of sets. Sets are similar to strings: you can pass them around but you can not modify them. Constant of type <cf>set int</cf> looks like <cf> [ 1, 2, 5..7 ]</cf>. As you can see, both simple values and ranges are permitted in sets. Sets of prefixes are special: you can specify which prefixes should match them by using <cf>[ 1.0.0.0/8+, 2.0.0.0/8-, 3.0.0.0/8{5,6} ]</cf>. 3.0.0.0/8{5,6} matches prefixes 3.X.X.X, whose prefix length is 5 to 6. 3.0.0.0/8+ is shorthand for 3.0.0.0/{0,8}, 3.0.0.0/8- is shorthand for 3.0.0.0/{0,7}. <tag/enum/ enumeration types are halfway-internal in the BIRD. You can not define your own variable of enumeration type, but some predefined variables are of enumeration type. Enumeration types are incompatible with each other, again, for your protection. <tag/bgppath/ bgp path is list of autonomous systems. <tag/bgpmask/ bgp mask is mask used for matching bgp paths (using <cf>path ~ / 2 3 5 ? / syntax </cf>). <cf/?/ is really serving in "any number of autonomous systems", but we did not want to use * because then it becomes too easy to write <cf>/*</cf> which is start of comment. <tag/clist/ community list. This is similar to set of pairs, except that unlike other sets, it can be modified. </descrip> <sect1>Operations Filter language supports common integer operations <cf>(+,-,*,/)</cf>, parentheses <cf/(a*(b+c))/, comparison <cf/(a=b, a!=b, a<b, a>=b)/. Special operators include <cf/˜/ for "in" operation. In operation can be used on element and set of that elements, or on ip and prefix, or on prefix and prefix or on bgppath and bgpmask or on pair and clist. Its result is true if element is in given set or if ip address is inside given prefix. Operator <cf/=/ is used to assign value to variable. <sect1>Control structures Filters support two control structures: if/then/else and case. Syntax of if/then/else is <cf>if <M>expression</M> then <M>command</M>; else <M>command</M>;</cf> and you can use <cf>{ <M>command_1</M>; <M>command_2</M>; <M>...</M> }</cf> instead of one or both commands. <cf>else</cf> clause may be omitted. <cf>case</cf> is similar to case from Pascal. Syntax is <cf>case <m/expr/ { else | <m/num_or_prefix [ .. num_or_prefix]/ : <m/statement/ ; [ ... ] }</cf>. Expression after <cf>case</cf> can be of any type that can be on the left side of ˜ operator, and anything that could be member of set is allowed before :. Multiple commands are allowed without {} grouping. If argument matches neither of : clauses, else: clause is used. (Case is actually implemented as set matching, internally.) Here is example that uses if and case structures: <code> case arg1 { 2: print "two"; print "I can do more commands without {}"; 3 .. 5: print "three to five"; else: print "something else"; } if 1234 = i then printn "."; else { print "*** FAIL: if 1 else"; } </code> <sect1>Route attributes Filter is implicitly passed route, and it can access its attributes, just like it accesses variables. <descrip> <tag>defined( <m>attribute</m> )</tag> returns TRUE if given attribute is defined. Access to undefined attribute results in runtime error. <tag/<m/prefix/ network/ network this route is talking about. <tag/<m/ip/ from/ who told me about this route. <tag/<m/ip/ gw/ what is next hop packets routed using this route should be forwarded to. <tag/<m/enum/ source/ what protocol told me about this route. This can have values such as <cf/RTS_RIP/ or <cf/RTS_OSPF_EXT/. </descrip> Plus, there are protocol-specific attributes, which are described in protocol sections. <sect1>Utility functions There are few functions you might find convenient to use: <descrip> <tag>print <m/expr/ [ <m/, expr .../ ]</tag> prints given expressions, useful mainly while debugging filters. <tag>quitbird</tag> terminates bird. Useful while debugging filter interpreter. </descrip> <sect>Protocols <sect1>BGP <sect1>Device <sect2>Introduction The Device protocol is not a real routing protocol as it doesn't generate any routes and only serves as a module for getting information about network interfaces from the kernel. Except for very unusual circumstances, you probably should always include this protocol in the configuration since almost all other protocol don't do anything if they are not provided with network interfaces. <sect2>Configuration <descrip> <tag>scan time <m/number/</tag> Time in seconds between two scans of the network interface list. On systems where we are notified about interface status changes asynchronously (such as newer versions of Linux), we need to scan the list only to avoid confusion by lost notifications, so the default time is set to a large value. </descrip> <sect2>Attributes As the Device protocol doesn't generate any routes, it cannot have any attributes. <sect2>Example <code> protocol device { scan time 10; # Scan the interfaces often } </code> <sect1>Direct <sect2>Introduction The Direct protocol is a simple generator of device routes for all the directly connected networks according to the list of interfaces provided by the kernel via the Device protocol. It's highly recommended to include this protocol in your configuration unless you want to use BIRD as a route server or a route reflector, that is on a machine which doesn't forward packets and only participates in distribution of routing information. <sect2>Configuration <descrip> <tag>interface <m/pattern/, <m/.../</tag> By default, the Direct protocol will generate device routes for all the interfaces available. If you want to restrict it to some subset of interfaces (for example if you're using multiple routing tables for policy routing and some of the policy domains don't contain all interfaces), just use this clause. </descrip> <sect2>Attributes Direct device routes don't contain any specific attributes. <sect2>Example <code> protocol direct { interface "-arc*", "*"; # Exclude the ARCnets } </code> <sect1>Kernel <sect1>OSPF <sect1>Pipe <sect1>Rip <sect2>Introduction Rip protocol (sometimes called Rest In Pieces) is simple protocol, where each router broadcasts distances to all networks he can reach. When router hears distance to other network, it increments it and broadcasts it back. Broadcasts are done in regular intervals. Therefore, if some network goes unreachable, routers keep telling each other that distance is old distance plus 1 (actually, plus interface metric, which is usually one). After some time, distance reaches infinity (that's 15 in rip) and all routers know that network is unreachable. Rip tries to minimize situations where counting to infinity is necessary, because it is slow. Due to infinity being 16, you can not use rip on networks where maximal distance is bigger than 15 hosts. You can read more about rip at <HTMLURL URL="http://www.ietf.org/html.charters/rip-charter.html">. <sect2>Configuration In addition to options generic to other protocols, rip supports following options: <descrip> <tag/authentication none|password|md5/ selects authentication method to use. None means that packets are not authenticated at all, password means that plaintext password is embedded into each packet, and md5 means that packets are authenticated using md5 cryptographic hash. If you set authentication to non-none, it is good idea to add <cf>passwords { }</cf> section. <tag>honor always|neighbor|never </tag>specifies, when should be routing table updates honored. (Always, when sent from host on directly connected network, or never.) </descrip> There are two options that can be specified per-interface. First is <cf>metric</cf>, with default one. Second is <cf>mode multicast|broadcast|quiet|nolisten|version1</cf>, it selects mode for rip to work in. If nothing is specified, rip runs in multicast mode. <cf>version1</cf> is currently equivalent to <cf>broadcast</cf>, and it makes rip talk at broadcast address even through multicast mode is possible. <cf>quiet</cf> option means that rip will not transmit periodic messages onto this interface and <cf>nolisten</cf> means that rip will talk to this interface but not listen on it. Following options generally override specified behavior from RFC. If you use any of these options, BIRD will no longer be RFC-compatible, which means it will not be able to talk to anything other than equally misconfigured BIRD. I warned you. <descrip> <tag>port <M>number</M></tag> selects IP port to operate on, default 520. (This is useful when testing BIRD, if you set this to address >1024, you will not need to run bird with UID==0). <tag>infinity <M>number</M></tag> select value of infinity, default 16. Bigger values will make protocol convergence even slower. <tag>period <M>number</M> </tag>specifies number of seconds between periodic updates. Default is 30 seconds. Lower number will mean faster convergence but bigger network load. <tag>timeouttime <M>number</M> </tag>specifies how old route has to be to be considered unreachable. Default is 4*period. <tag>garbagetime <M>number</M> </tag>specifies how old route has to be to be discarded. Default is 10*period. </descrip> <sect2>Attributes RIP defines two route attributes: <descrip> <tag>int <cf/rip_metric/</tag> RIP metric of the route (ranging from 0 to <cf/infinity/). When routes from different RIP instances are available and all of them have the same preference, BIRD prefers the route with lowest <cf/rip_metric/. <tag>int <cf/rip_tag/</tag> RIP route tag: a 16-bit number which can be used to carry additional information with the route (for example, an originating AS number in case of external routes). </descrip> <sect2>Example <code> protocol rip MyRIP_test { debug all; port 1520; period 7; garbagetime 60; interface "eth0" { metric 3; mode multicast; } "eth1" { metric 2; mode broadcast; }; honor neighbour; passwords { password "ahoj" from 0 to 10; password "nazdar" from 10; } authentication none; import filter { print "importing"; accept; }; export filter { print "exporting"; accept; }; } </code> <sect1>Static <sect2>Introduction The static protocol doesn't communicate with other routers in the network, but instead it allows you to define routes manually which is often used for specifying how to forward packets to parts of the network which don't use dynamic routing at all and also for defining sink routes (i.e., those telling to return packets as undeliverable if they are in your IP block, you don't have any specific destination for them and you don't want to send them out through the default route to prevent routing loops). There are three types of static routes: `classical' routes telling to forward packets to a neighboring router, device routes specifying forwarding to hosts on a directly connected network and special routes (sink, blackhole etc.) which specify a special action to be done instead of forwarding the packet. When the particular destination is not available (the interface is down or the next hop of the route is not a neighbor at the moment), Static just uninstalls the route from the table its connected to and adds it again as soon as the destinations becomes adjacent again. <sect2>Configuration The Static protocol has no configuration options. Instead, the definition of the protocol contains a list of static routes which can contain: <descrip> <tag>route <m/prefix/ via <m/ip/</tag> Static route through a neighboring router. <tag>route <m/prefix/ via <m/"interface"/</tag> Static device route through an interface to hosts on a directly connected network. <tag>route <m/prefix/ drop|reject|prohibit</tag> Special routes specifying to drop the packet, return it as unreachable or return it as administratively prohibited. </descrip> <sect2>Attributes Static routes have no specific attributes. <sect2>Example <code> protocol static { table testable; # Connect to non-default routing table route 0.0.0.0/0 via 62.168.0.13; # Default route route 62.168.0.0/25 reject; # Sink route route 10.2.0.0/24 via "arc0"; # Secondary network } </code> <sect>Getting more help This is really last section of this file, should give pointers to programmers documentation, web pages mailing lists and similar stuff. </article>