- Bird-users - Mailing Lists

RIP TTL security Ubuntu bug ?
by olivier a 07 Apr '15

07 Apr '15

Hi Trying to use 'ttl security enable' with RIP protocol in BIRD, I get : /etc/bird.conf, line 30, syntax error It happens on three machines : . Ubuntu 14.04 LTS / Bird installed by repository . Ubuntu 14.04 LTS / Bird installed with 1.4.5 sources compile . Raspbian OS, Bird installed with 1.4.5 sources compile I tried with different syntaxes ( enable, yes, on, .. ) with same pb Did I miss a compile option ? ttl security if functioning OK with BIRD/BGP Thanks for any insight :-)

2 2

BIRD / BGP / IPv6 / source address directive : auto vs link local vs GUA
by olivier a 04 Apr '15

04 Apr '15

Hi ! I'm learning BGP and I'm really puzzled about the 'source address' directive in the context of IPv6. I've seen some people explicit a GUA source address ( example : source address 2001:db8:0:0::1; ) Reading the documentation, watching the debug logs and playing with bird6c didn't help me understand. Leaving to auto ( no directive ) was functioning ok, but my BGP setup is rather simple. What is better ? No directive, link local or Global Unicast Address ? Thank you for any clue :-)

3 3

Modifying function called in filter does not take effect
by Eric Cables 01 Apr '15

01 Apr '15

I'm hoping someone can tell me whether this behavior is expected or not. I am using BIRD as a RR, and running into a behavior where modifying a policy, within a function called by a filter, requires a hard BGP session reset or stop/start of BIRD, rather than a 'configure soft.' Route Refresh is supported by the RR client. I am using a unique filter for each RR client; each filter calls a function that is shared amongst all filters -- basically matching routes that all RR clients should receive. This was first instantiated with a single BGP community match in the function, and a 'configure soft' worked fine. I then went back and added a second if statement to the function, expecting 'configure soft' to put the new policy into effect, but what I noticed is that the only way for RR clients to receive the newly accepted routes was to hard down/up the BGP session. Anecdotally, a 'show route filter router1_out' (after 'configure soft') did match the routes I was expecting the RR client to receive, but until the BGP session was hard cleared to the RR client, they were not advertised by BIRD. Here is the relevant configuration (I've bolded the new configuration that was added, which did not result in an advertisement to the RR client): # RR client template template bgp rr_client { description "Template for route reflector clients"; local as 65412; hold time 180; source address 2.2.2.2 ; rr client; rr cluster id 1.1.1.1; import all; export none; add paths; } # Globally reflected routes function function global_routes() { # Match 65000-65499,100-200 if bgp_community ~ [(65000..65499,100..200)] then accept; * # Match 65000-65499,300-400* * if bgp_community ~ [(65000..65499,300..400)] then accept;* } # Outbound filter for router1 filter router1_out { global_routes(); reject; } protocol bgp 'router1' from rr_client { neighbor 1.2.3.4 as 65412; import all; export filter router1_out; } As mentioned, the output of 'show route filter router1_out' (after a 'configure soft') did return the expected routes, but they were not advertised to the 'router1' RR client until the session was hard cleared. Is this expected behavior? -- Eric Cables

2 1

Is it good practice to use same router id for bird and bird6 ?
by olivier a 28 Mar '15

28 Mar '15

Hi ! Is it good practice to use same router id for bird and bird6 ? I'm wondering if there are some drawbacks or pitfalls ? I'm using RIP, OSPF and BGP Thanks for any advice :-)

5 6

RIP and directly connected routes
by olivier a 28 Mar '15

28 Mar '15

I'm studying RIP and have a really hard time understanding the rationals behind to automatically pulling directly connected routes. If I connect 4 RIP routers in a circle and neither is advertising it's connected routes, it's going to be very quiet ... Please help me understand :-) Secondly, I use 'protocol direct' to have these connected routes pulled from kernel tables to RIP and advertised. Is that the right way to do it ? Lastly, these 'protocol direct' routes override the split horizon rule. Is there a way to prevent an interface to multicast its own subnet ? Thanks, I 'm just discovering BIRD, I may be asking clumsy questions ;-)

2 3

New version, v0.5, of BPGSEC implementation for BIRD
by Michael Baer 24 Mar '15

24 Mar '15

Hi all, I wanted to announce a new, v0.5, version of BGPsec protocol supporting code for BIRD. It's been updated to work with v1.4.5 BIRD and is available as a patch or as a full source tarball at: http://bgpsec.tislabs.com This is an ongoing project following the IETF's SIDR Working Group's BGPsec implementation. This version has been updated to match the current changes in the draft BGPsec protocol specification. Notably, the code also includes a client, bgpsec-bird-client, that uses RTRlib (http://rpki.realmv6.org/trac/) to download RPKI data from a Resource Public Key Infrastructure (RPKI) to Router Protocol server and populate BIRD's ROA tables. It will also download router keys from the RPKI to Router protocol server. SIDR working group and protocol links: http://datatracker.ietf.org/wg/sidr/charter/ http://datatracker.ietf.org/doc/draft-ietf-sidr-bgpsec-protocol/ http://tools.ietf.org/html/rfc6810 (RPKI to Router Protocol) The code is still at the experimental/testing stage and should not be use used for production services. If any one wants to play with the code, please feel free to email me any questions or bug reports. Thanks, Mike -- Michael Baer PARSONS baerm(a)tislabs.com

1 0

Force BGP add-path configuration option
by Eric Cables 20 Mar '15

20 Mar '15

I realize that BGP add-path is a negotiated capability, but I'm running into an instance where I am attempting to send multiple copies of the same prefix, with different BGP attributes (next hop, etc.), but since the receiving router does not support add path BIRD is only sending its "best" prefix. Would it be possible to add a "add paths force" configuration option? What potential implications would this cause for the receiving device that is unaware of this capability, and likely will never have support for it? -- Eric

1 0

on mrt support
by Evelio VILA 19 Mar '15

19 Mar '15

Hi, Does bird dump mrt rib entries? as per rfc rfc6396 ? i see it has support for bgp messages and states but not the rib. How exactly does the dump routes command works? thanks Evelio Vila

2 1

BIRD ignored routes
by Nat J 12 Mar '15

12 Mar '15

Hi All, I’m having an issue with bird ignoring some prefixes it’s receiving. I'm new to bird and may be overlooking something simple. The config is essentially this [1] Exabpg1:AS_X < — > BIRD:AS_Y (route-server) [2] Exabgp2:AS_X < — > BIRD:AS_Y (route-server) [3] BIRD:AS_Y < — > BIRD:AS_Z [1] & [2] behave as expected, however when the route-server exports to the 2nd BIRD instance the routes are ignored. The debug file unfortunately makes no mention of the routes being ignored. Any suggestions on how I can find the cause for those being ignored. Attached are the configs and outputs from BIRD Thanks -nat

2 2

BGP export connected or static prefixes
by Anand Buddhdev 09 Mar '15

09 Mar '15

Dear BIRD users, I'm setting up BIRD 1.4.5 as a router for an AS112 anycast instance. I've got the router running and creating peering sessions with peers, and also importing prefixes from them into the kernel. However, it's not exporting my prefix, and I'm afraid I don't understand why. I'm quite new to BIRD, so please bear with me if my question looks naive. I would appreciate any clues. My configuration looks like this: list bgp address <myaddr>; router id <myaddr>; tabe anycast; filter as112 { if net = 192.75.48.0/24 then accept; reject; } protocol direct { interface "dummy0*"; } protocol device { } protocol kernel { export all; kernel table 10; table anycast; } protocol bgp { export filter as112; local <myaddr> as 112; neighbor <peeraddr> as <peeras>; table anycast; }

2 6