- Bird-users - Mailing Lists

Question about BGP advertisement-interval in BIRD
by Kim Charles 10 Aug '20

10 Aug '20

Hi, I'm currently using BIRD version 1.6.3 and I'm looking for any equivalent in BIRD for FRR BGP advertisement-interval option as below. It would be appreciated if you can help me on this. [no] neighbor PEER advertisement-interval (0-600) Setup the minimum route advertisement interval(mrai) for the peer in question. This number is between 0 and 600 seconds, with the default advertisement interval being 0. Best Regards, Charles -- Charles Kim Stony Brook MS-TM / aSSIST MBA / 7xCCIE#12303 Mobile : +82-10-8918-9777 LinkedIn : https://www.linkedin.com/in/charles-kim-5b50a02b/

1 0

Status of implementing multicast routing
by Blažej Krajňák 06 Aug '20

06 Aug '20

Hello, I was unable to find more details. Is there any plan to implement multicast routing (SSM, DNS support)? Thanks, Blažej

4 5

static route - policy based routing in Bird,
by Farhad 05 Aug '20

05 Aug '20

Hello, I am new to bird, there is something which I don't know how to achieve using bird. Is there a way in which I can forward traffics originated from a specific network to some designated interfaces? protocol kernel { ipv4 { table tabl1; export all; }; kernel table 1; } protocol kernel { ipv4 { table tabl2; export all; }; kernel table 2; } protocol static { ipv4 { table table1; }; route 0.0.0.0/0 via 10.7.3.249 bfd via 10.9.3.251 bfd; } protocol static { ipv4 { table table2; }; route 0.0.0.0/0 via 10.8.3.244 bfd; } In linux i can do something like this: *ip rule add type unicast from 1.0.0.1 table 1* *ip rule add type unicast from 2.0.0.1 table 2* Is there a way these rules can be achieved using filter or Pipe in Bird? I am curious whether this can be done in Bird without depending on 'ip rule' of linux. Regards, Farhad

1 0

Analyzing AS PATH in packets
by Julien Sansonnens 04 Aug '20

04 Aug '20

Dear all, using bird 2.0.7 I'm a bit confused with the output of "show route export". I'm exporting a full table to a downstream. I'm expecting to see my own AS as the first AS in AS PATH but it will not show in "BGP.as_path" If I prepend it, then it appears on the output. I don't know if it's related to display, or if my AS is really not added to the PATH. Perhaps it is added later in the process, before leaving my router ? How can I check this? I tried with "debug", but to my knowledge, it is not possible to display route details. I thought about doing tcpdump, but I haven't found out how to interpret the result correctly and see the AS PATH. Would you have a clue on how to see exactly what is being exported on an interface? thanks a lot Julien

2 1

Bird 2 debugging, connections getting rejected
by Skyler Mäntysaari 03 Aug '20

03 Aug '20

Hello, I'm trying to get IBGP session to work between Bird 2 and FFRouting (EdgeRouter). They are using Wireguard for the link, and I can ping them fine. BGP port is also open on both. What's happening here? Errors I'm getting: -------- 2020-08-03 23:51:56.870 <TRACE> ibgp_tku: Incoming connection from 2001:67c:1104:1a00::1 (port 46843) rejected -------- Full bird config: ---------------------------------------------------------------------------------------------------------------------------------------- # Configure logging #log syslog all; #log "/var/log/bird.log" { debug, trace, info, remote, warning, error, auth, fatal, bug }; log "/var/log/bird.log" all; #log syslog all; debug protocols { events, states }; # Set router ID. It is a unique identification of your router, usually one of # IPv4 addresses of the router. It is recommended to configure it explicitly. router id 45.92.40.5; # Turn on global debugging of all protocols (all messages or just selected classes) #debug protocols all; # debug protocols { events, states }; # Turn on internal watchdog # watchdog warning 5 s; # watchdog timeout 30 s; # You can define your own constants define my_asn = 213021; # define my_addr = 198.51.100.1; # Tables master4 and master6 are defined by default ipv4 table master4; ipv6 table master6; # Define more tables, e.g. for policy routing or as MRIB ipv4 table mrib4; ipv6 table mrib6; # The Device protocol is not a real routing protocol. It does not generate any # routes and it only serves as a module for getting information about network # interfaces from the kernel. It is necessary in almost any configuration. protocol device { scan time 5; } protocol direct DIRECT { ipv4; ipv6; } protocol kernel KERNEL4 { ipv4 { export all; import none; }; } protocol kernel KERNEL6 { ipv6 { export all; import none; }; } protocol static STATIC6 { ipv6; route 2a0b:a700:3::2/128 via 2a07:22c0:c100::; #route 2001:67c:1104:1a00::1 dev wg0; #route 2a01:4f9:2b:f800::2 via 2a07:22c0:c100::; #route 2a01:4f8::/29 via 2001:7f8:33::a105:8299:1; } filter internal_nets_v6 { if net ~ [ 2a0e:b107:6ec::/48,2a0e:8f02:2010::/44,2001:67c:1104::/48 ] then accept; reject; } filter no_export_internal_nets { if net ~ [ 2a0e:b107:6ec::/48,2a0e:8f02:2010::/44,2001:67c:1104::/48 ] then { bgp_community.add((65535:65281)); accept; } reject; } filter internal_route { if net ~ [ 2a07:22c0:c100::/64 ] then reject; accept; } template bgp T_SELF { local 2001:67c:1104:1a00::2 as 213021; ipv6 { import filter internal_nets_v6; export filter internal_route; }; } protocol bgp ibgp_tku from T_SELF { neighbor 2001:67c:1104:1a00::1 as 213021; #neighbor fe80::21b:21ff:febb:5db0%wg0 as 213021; direct; ipv6 { import filter internal_nets_v6; export all; next hop self; }; } protocol bgp ibgp_de from T_SELF { local 2001:67c:1104:1a01::1 as 213021; neighbor 2001:67c:1104:1a01::2 as 213021; direct; ipv6 { import all; export all; next hop self; }; } ---------------------------------------------------------------------------------------------------------------------------------------- EdgeRouter side: ---------------------------------------------------------------------------------------------------------------------------------------- address-family { ipv6-unicast { prefix-list { export ANNOUNCE-THESE } soft-reconfiguration { inbound } } } description "Frankfurt, DE" remote-as 213021 update-source 2001:67c:1104:1a00::1 ----------------------------------------------------------------------------------------------------------------------------------------

1 0

Re: Bird 2.0 Install
by Krish R. [VNP DIGITAL LTD.] 03 Aug '20

03 Aug '20

Hi, I had the same issue on Ubuntu 18.04 LTS. I ran the following commands wget http://archive.ubuntu.com/ubuntu/pool/main/i/init-system-helpers/init-syste… (http://archive.ubuntu.com/ubuntu/pool/main/i/init-system-helpers/init-syste…) dpkg -i init-system-helpers_1.57_all.deb then run apt-get install bird2. Thanks August 3, 2020 10:57 AM, "Jason Kopacko" <jason(a)kopacko.com (mailto:jason@kopacko.com?to=%22Jason%20Kopacko%22%20<jason(a)kopacko.com>)> wrote: It's not simple. I have been trying everything I can find via Google because packages just aren't available or something. root@#####:~# apt-get install bird2 Reading package lists... Done Building dependency tree Reading state information... Done Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: The following packages have unmet dependencies: bird2 : PreDepends: init-system-helpers (>= 1.56~) but 1.51 is to be installed E: Unable to correct problems, you have held broken packages. Sorry if I come off as rude, I just fed up with trying to install this today. Sent from my Verizon Wireless 4G LTE Droid Turbo 2. On Sun, Aug 2, 2020 at 7:50 PM Kr <kr(a)vnpdigital.com (mailto:kr@vnpdigital.com)> wrote: Hi, It is quite simple. You can run apt-get install bird2. If you run into any issues, don’t hesitate to contact me. Cheers. Krish On Aug 3, 2020 at 10:23 am, <Jason Kopacko (mailto:jason@kopacko.com)> wrote: I am beginning to give up. As a very inexperienced Ubuntu user, I am trying to install bird 2.x. But cannot seem to figure out how to use apt-get to install anything above bird 1.6. What am I doing wrong and why is this sooooo hard? Sent from my Verizon Wireless 4G LTE Droid Turbo 2. (https://vnpdigital.com)

2 1

How to pipe ipv6 channel into ipv6 channel or possible alternatives
by Nick Cao 03 Aug '20

03 Aug '20

I'm trying to use babel (namely, with the ipv6 sadr channel) together with bgp, and thus I got route from both channels to be installed into kernel. But the limitation of kernel protocol dictates that I seem to have no choice other than separate the routes into different kernel routing tables, then use ip rules to take them into effect. However I found some interesting discussion in the mailing list (https://bird.network.cz/pipermail/bird-users/2017-May/011277.html) indicating that we can convert (or pipe) ipv6 channel into ipv6 sadr channel, which, sadly still has't made it's way into bird as far as I know. I'd like to ask whether there's some way to mimic that sort of behavior, or is there any alternative to that so that ipv6 sadr channel can play nicely with the ipv6 channel. ---------------------------------------- Nick Cao <nickcao(a)nichi.co>

1 0

Automated builds of Bird 2 for mips
by Skyler Mäntysaari 03 Aug '20

03 Aug '20

Hello, I was wondering that does anyone have an existing automated CI pipeline to build a Bird 2 .deb package? If not, got any pointers on the QEMU image run with Travis CI, Circle CI, Gitlab CI or some other public/open source CI?

1 0

Bird 2.0 Install
by Jason Kopacko 03 Aug '20

03 Aug '20

I am beginning to give up. As a very inexperienced Ubuntu user, I am trying to install bird 2.x. But cannot seem to figure out how to use apt-get to install anything above bird 1.6. What am I doing wrong and why is this sooooo hard? Sent from my Verizon Wireless 4G LTE Droid Turbo 2.

2 2

BIRD trying to reinsert existing kernel routes, netlink issue?
by Sasha Romijn 02 Aug '20

02 Aug '20

Hello, I’m running into a rather complex issue where BIRD tries to reinsert routes into the kernel routing table that are already in there. I’m at the point of suspecting a netlink bug in the 5.x linux kernel. However, I may be entirely off, and would like to hear what people with more BIRD and/or netlink expertise think about it. My setup is bird 2.0.7 on Ubuntu 20.04. I’ve been able to reproduce this with the default 5.4.0-42-generic kernel as well as the ubuntu mainline 5.7.0-050700-generic kernel. This bird instance has an IBGP peering with another bird 2.0.7 instance, which sends an IPv4 full table. The IBGP peering is set up over a GRE tunnel. There is no other routing software running. I am seeing this issue on another server too, ubuntu 20.04 with 5.4.0-42-generic. A third one with ubuntu 18.04 and a 4.15.0-20-generic kernel, is fine. The problem presents as follows: a few thousand routes fail insertion into the kernel routing table, marked by bird as ‘!’: ------------------------- # birdc show route table master4|grep \!|wc -l 5838 ------------------------- The number varies, but is in the order of a few thousand. However, all routes seem to be in the kernel table: ------------------------- # birdc show route count BIRD 2.0.7 ready. 788748 of 788748 routes for 788747 networks in table master4 0 of 0 routes for 0 networks in table master6 Total: 788748 of 788748 routes for 788747 networks in 2 tables # ip route|grep bird|wc -l 788748 ------------------------- When I wait a few minutes and check again, the set of failed prefixes is almost entirely different. Apparently, the previously failed prefixes have been “fixed”, but now the same problem appears with different ones. When I check a particular sample, the route really is in the kernel routing table already: ------------------------- # birdc show route 81.200.176.0/20 BIRD 2.0.7 ready. Table master4: 81.200.176.0/20 unicast [fra1 14:47:20.892] ! (100) [AS50664i] via 10.195.6.1 on tun-fra1 # ip route|grep 81.200.176.0/20 81.200.176.0/20 via 10.195.6.1 dev tun-fra1 proto bird metric 32 # birdc show route 81.200.176.0/20 BIRD 2.0.7 ready. Table master4: 81.200.176.0/20 unicast [fra1 14:47:20.892] ! (100) [AS50664i] via 10.195.6.1 on tun-fra1 ------------------------- The logs are repeats of: ------------------------- Aug 01 09:50:23 fra2 bird[83586]: Netlink: File exists Aug 01 09:50:23 fra2 bird[83586]: Netlink: File exists Aug 01 09:50:23 fra2 bird[83586]: Netlink: File exists Aug 01 09:50:23 fra2 bird[83586]: Netlink: File exists Aug 01 09:50:23 fra2 bird[83586]: Netlink: File exists Aug 01 09:50:23 fra2 bird[83586]: ... Aug 01 09:50:23 fra2 bird[83586]: I/O loop cycle took 5110 ms for 1 events ------------------------- Seemingly, bird thinks routes are not yet in the kernel routing table, tries to insert them, which fails because they are already there. Later scans move the problem to different prefixes. If I remove the tunneled IBGP peering, and set up an upstream peering on a directly connected non-tunneled peer, everything works fine. If I enable both the tunneled IBGP peering and the non-tunneled upstream, the issue does appear. The routes that are affected by the failed insertion attempts are then from both peers. The prefixes that are affected are almost entirely cases where multiple prefixes exist in the DFZ with the same network address, but different lengths. Above I mentioned seeing 81.200.176.0/20 at some point, and 81.200.176.0/24 is also in the DFZ. From two samples of failed prefixes at different points in time, about 98% are prefixes of this kind, where the DFZ contains the same network address with a different length too. This distribution is not reflected in the whole v4 full table, suggesting that routes that meet this case, are most likely to fail insertion. I have done some deeper digging with strace. https://p.6core.net/p/RtQ7xSSYd560i7FioG78ebsi is an strace of BIRD starting and loading the v4 full table from the IBGP peer. I have filtered this for “45.189.104.0” “GETROUTE” to trim down. I have the full strace, but it’s 8GB: 1. Initially, bird does GETROUTE a few times, before the BGP session is established. 2. Near the start, at 09:06:33.208648 and 09:06:33.209003, the BGP session seems established, and 45.189.104.0/24 and 45.189.104.0/22 are inserted correctly. (Acknowledgement not in paste as it does not contain the prefix.) 3. At 09:06:39.774454 bird sends to birdc that 45.189.104.0/24 is best route and inserted correctly. 4. At 09:06:49.915838, bird sends a GETROUTE. Only the reply lines that contain “45.189.104.0” are in the paste. Apparently, only 45.189.104.0/24 is included in the reply. 45.189.104.0/22 does not seem to appear in the reply to GETROUTE. 5. At 09:07:08.455767, bird tries to insert 45.189.104.0/22 into the kernel routing table. This fails, because it already exists. 6. At 09:07:43.144207 bird replies to birdc that 45.189.104.0/22 failed to insert. 7. At 09:07:49.874282, bird does another GETROUTE. Both the /22 and the /24 are included in the response. 8. At 09:08:25.623269, bird tries to update 45.189.104.0/22 which succeeds. Note the different flags from 09:07:08.455767, specifically NLM_F_EXCL vs NLM_F_REPLACE, presumably because bird is aware there is an existing route. 9. Future GETROUTEs return both the /24 and /22, bird does nothing. Working theory: netlink GETROUTE in at least some 5.x kernels may not return all routes, when at least some routes in the table have a next hop that is a tunnel interface, and this is almost entirely contained to cases where multiple prefixes exist with the same network address. Thoughts? And if that theory is correct, can we work around it in BIRD? A few things I’ve changed that had no effect: - Changing the tunnel type. - Changing the peering from IBGP to EBGP. - Using IPv6 addresses for the tunnel endpoints, and removing the static route in the config below. This is the full config I’m currently using: ------------------------- log syslog all; router id 141.98.136.36; protocol kernel { scan time 20; ipv4 { export all; import none; }; } # never route the tunnel through the tunnel protocol static tunnel_endpoints { ipv4; route 45.12.69.14/32 via 141.98.136.33; } protocol device { } # ibgp peering over GRE tunnel, ipv4 full table protocol bgp fra1 { local as 213279; source address 10.195.6.2; neighbor 10.195.6.1 as 213279; direct; ipv4 { import all; export none; }; } ------------------------- Sasha

3 3