July 2020 - Bird-users - Mailing Lists

Static routes, preferences and ordering
by William 24 Jul '20

24 Jul '20

Hi All, Discovered something interesting in bird2 2.0.7 (via bird2 package on Centos7, 2.0.7-1.el7) when retrying to use statics to override a default route coming via BGP. The following block would not see the higher preference route added (not even listed in 'show route protocol WANSTATICS'), but when commented out and config reloaded the gw via 172.16.4.250 would vanish, then come back when I restarted WANSTATICS. protocol static WANSTATICS { ipv4 { table WAN; }; route 0.0.0.0/0 via 172.16.1.254 { preference = 110; # override upstream }; route 0.0.0.0/0 via 172.16.4.250 { preference = 90; # prefer one to come from upstream via BGP }; check link on; } It was working on another sibling router node, but the two route blocks were reversed: protocol static WANSTATICS { ipv4 { table WAN; }; route 0.0.0.0/0 via 172.16.4.250 { preference = 90; # prefer one to come from upstream via BGP }; route 0.0.0.0/0 via 172.16.1.254 { preference = 110; # override upstream }; check link on; } Lo and behold, switch them around on the problematic node and suddenly everything behaves, 'show route protocol WANSTATICS' lists the new route and it propagates into the routing table too. There's nothing that I could see in the documentation to explain this, wondering if anyone else has seen something similar? Regards, William

2 1

Re: [BFD] BFD is Not working after system reboot.
by seunghwan hwang 23 Jul '20

23 Jul '20

Hello K. What does "disabled" setting mean? This is my configuration ================================== protocol bfd { multihop { min rx interval 300 ms; min tx interval 300 ms; multiplier 3; }; } protocol bgp as1 { import all; export all; local ~ as ~; neighbor ~ as ~; enable route refresh off; hold time 90; bfd on; } protocol bgp as2 { import all; export all; local ~ as ~; neighbor ~ as ~; enable route refresh off; hold time 90; bfd on; } ================================== Please check it. Regards. 2020년 7월 20일 (월) 오후 7:48, Kees Meijs | Nefos <kees(a)nefos.nl>님이 작성: > Hello Seunghwan, > > Haven't you got a "disabled" setting in de BIRD configuration file in > the protocol portion? > > Regards, > Kees > > On 20-07-2020 03:54, seunghwan hwang wrote: > > > > I've already tried capturing tcpdump. > > BIRD's BFD neighbor was captured sending packets, but the packets sent > > by BIRD were not captured. > > In this state, BIRD's BFD packets were captured after restarting BFD > > protocol. > > The BFD packet was captured Only after restarting the BFD protocol . > > Is there an issue with BFD Protocol? > > Please Check it. > > > >

1 0

import bgp ipv6 routes issue
by Mazur, Dariusz 22 Jul '20

22 Jul '20

Hello Bird Users, Hope you will be able to help me. I have 8 bgp peers which use the same bgp template with below import filter: filter IMPORT__SPINE__TO__LEAF { bgp_community.add(LOOP_DETECTION); gw = bgp_next_hop; -> I use it to set ipv6 bgp next hop to global ipv6 address (not link local) if (net = 0.0.0.0/0 || net = ::/0) then { reject; } … else { reject; } } 1.First Peer -> this look good r14.spine101.fra02.fab> bird show protocols all 2a02:26f0:3700:1101::1__r01.leaf101 BIRD 2.0.7 ready. Name Proto Table State Since Info 2a02:26f0:3700:1101::1__r01.leaf101 BGP --- up 09:58:55.848 Established … Channel ipv6 State: UP Table: master6 Preference: 100 Input filter: IMPORT__SPINE__TO__LEAF Output filter: EXPORT__SPINE__TO__LEAF Routes: 27 imported, 4 exported, 25 preferred -> I expect to import 27 routes Route change stats: received rejected filtered ignored accepted Import updates: 27 0 0 0 27 Import withdraws: 0 0 --- 0 0 Export updates: 15846 50 15791 --- 5 Export withdraws: 11 --- --- --- 0 BGP Next hop: 2a02:26f0:3700:1101:: 2.Second Peer -> this does not look good: r14.spine101.fra02.fab> bird show protocols all 2a02:26f0:3700:1102::1__r02.leaf101 BIRD 2.0.7 ready. Name Proto Table State Since Info 2a02:26f0:3700:1102::1__r02.leaf101 BGP --- up 09:58:56.676 Established Channel ipv6 State: UP Table: master6 Preference: 100 Input filter: IMPORT__SPINE__TO__LEAF Output filter: EXPORT__SPINE__TO__LEAF Routes: 4 imported, 4 exported, 0 preferred -> bird imports only 4 routes Route change stats: received rejected filtered ignored accepted Import updates: 27 0 23 0 4 -> bird reject 23 routes After I reload this protocol all looks good and bird import 27 routes In logs I see “Invalid gw address”, therefore I suspect issue with “gw = bgp_next_hop” 2020-07-22 09:58:56.801 <ERR> filters, line 169: Invalid gw address 2020-07-22 09:58:56.801 <ERR> filters, line 169: Invalid gw address 2020-07-22 09:58:56.801 <ERR> filters, line 169: Invalid gw address 2020-07-22 09:58:56.801 <ERR> filters, line 169: Invalid gw address 2020-07-22 09:58:56.801 <ERR> filters, line 169: Invalid gw address I am able to reproduce issue on more routers, please find below out from another router: r16.spine101.fra02.fab> show bgp summary -> it parse “show protocols all” IPv6 Peers: 10 Down: 0 Peer AS State (Im/Filt/Exp/Pref) Up/Down Protocol 2a02:26f0:3700:211:: 20940 Established (8480/0/35/8480) 00:36:46 2a02:26f0:3700:211::__r01.border101 2a02:26f0:3700:311:: 20940 Established (7131/0/34/7) 00:36:48 2a02:26f0:3700:311::__r02.border101 2a02:26f0:3700:1301::1 20940 Established (28/0/4/26) 00:36:51 2a02:26f0:3700:1301::1__r01.leaf101 2a02:26f0:3700:1302::1 20940 Established (4/0/4/0) 00:36:49 2a02:26f0:3700:1302::1__r02.leaf101 -> 4 routes are imported but should be 28 2a02:26f0:3700:1303::1 20940 Established (28/0/4/1) 00:36:53 2a02:26f0:3700:1303::1__r03.leaf101 2a02:26f0:3700:1304::1 20940 Established (4/0/4/0) 00:36:50 2a02:26f0:3700:1304::1__r04.leaf101 2a02:26f0:3700:1305::1 20940 Established (28/0/4/1) 00:36:47 2a02:26f0:3700:1305::1__r05.leaf101 2a02:26f0:3700:1306::1 20940 Established (28/0/4/1) 00:36:46 2a02:26f0:3700:1306::1__r06.leaf101 2a02:26f0:3700:1307::1 20940 Established (28/0/4/1) 00:36:47 2a02:26f0:3700:1307::1__r07.leaf101 2a02:26f0:3700:1308::1 20940 Established (28/0/4/1) 00:36:46 2a02:26f0:3700:1308::1__r08.leaf101 After protocol reload all looks good. Did you face similar problem? Regards, Dariusz

2 2

[PATCH] Enable class E address space via --enable-class-e flag
by Lincoln Dale 21 Jul '20

21 Jul '20

This change enables Bird to establish BGP sessions to devices in class E address space. Without this change, prefixes in class e announced via BGP are unconditionally ignored, e.g: $ bird -d bird: Started bird: Ignoring bogus route 241.1.2.3/32 received via <neighbor> After this change, if configuration flag --enable-class-e is provided, class E / bogons can still be ignored/filtered via routing policy but aren't otherwise explicitly marked as invalid. Marking class E addresses as bogus was previously discussed in Sep 2017 with no patch and no proposed change of behaviour ( https://bird.network.cz/pipermail/bird-users/2017-September/011558.html) Signed-off-by: lincoln dale <ltd(a)interlink.com.au> --- diff --git a/configure.ac b/configure.ac index 40f021a..3f22fa5 100644 --- a/configure.ac +++ b/configure.ac @@ -48,6 +48,12 @@ AC_ARG_ENABLE([mpls-kernel], [enable_mpls_kernel=try] ) +AC_ARG_ENABLE([class-e], + [AS_HELP_STRING([--enable-class-e], [allow class E addresses to be used])], + [enable_class_e=yes], + [enable_class_e=no] +) + AC_ARG_WITH([protocols], [AS_HELP_STRING([--with-protocols=LIST], [include specified routing protocols @<:@all@:>@])], [], @@ -391,6 +397,10 @@ if test "$enable_debug" = yes ; then fi fi +if test "$enable_class_e" = yes ; then + AC_DEFINE([ENABLE_CLASS_E], [1], [Define to 1 to enable support for class E addresses.]) +fi + CLIENT=birdcl CLIENT_LIBS= if test "$enable_client" = yes ; then @@ -454,6 +464,7 @@ AC_MSG_RESULT([ Debugging: $enable_debug]) AC_MSG_RESULT([ POSIX threads: $enable_pthreads]) AC_MSG_RESULT([ Routing protocols: $protocols]) AC_MSG_RESULT([ Kernel MPLS support: $enable_mpls_kernel]) +AC_MSG_RESULT([ Allow Class E addresses: $enable_class_e]) AC_MSG_RESULT([ Client: $enable_client]) rm -f $objdir/.*-stamp diff --git a/lib/ip.c b/lib/ip.c index 2d19516..143ef01 100644 --- a/lib/ip.c +++ b/lib/ip.c @@ -103,6 +103,12 @@ ip4_classify(ip4_addr ad) if (a == 0xffffffff) return IADDR_BROADCAST | SCOPE_LINK; +#ifdef ENABLE_CLASS_E + /* consider class E to be valid */ + if (b >= 0xf0) + return IADDR_HOST | SCOPE_UNIVERSE; +#endif + return IADDR_INVALID; } ---

2 2

Repository Mirror
by Julian Fölsch 20 Jul '20

20 Jul '20

Good day, In our environment we have Debian machines we want to run bird2 on. However Debian only offers bird1 in their repositories. I noticed that you offer a repository at bird.network.cz/debian but was unable to find any information on mirroring guidelines. Poking bird.network.cz for a running rsync server also yielded no results. It would be great if you could offer rsync access so syncing a mirror would be easy to do. Or if you expect to mirror the content by different means, please tell me. Kind regards, Julian Fölsch -- Julian Fölsch Arbeitsgemeinschaft Dresdner Studentennetz (AG DSN) Telefon: +49 351 271816 69 Mobil: +49 152 22915871 Fax: +49 351 46469685 Email: julian.foelsch(a)agdsn.de Studierendenrat der TU Dresden Helmholtzstr. 10 01069 Dresden

1 0

Bird2 for Debian Jessie?
by Skyler Mäntysaari 16 Jul '20

16 Jul '20

Hello there, Where can I find Bird2 for Jessie? If it's indeed required to make my own debian package for it, instructions on how? The repo at https://bird.network.cz/debian/ only has 1.6.x. Best regards, Skyler Mäntysaari

4 8

Getting started with OSPFv3 and Bird 2.0.7
by Skyler Mäntysaari 15 Jul '20

15 Jul '20

Hello List, I'm trying to wrap my head around how I'm supposed to get OSPv3 working though Wireguard interfaces to VyOS. Goal is to export my own prefixes from VyOS to Bird. Am I even on the right track about doing this? Router running Bird, is at the IX. Router running VyOS is tunneled though Wireguard. Bird config, that I tried: protocol ospfv3 main { import all; export all; area 0.0.0.0 { interface "wg0" { hello 1; dead 4; type broadcast; }; }; } That failed due to: "bird: /etc/bird/bird.conf:88:10 syntax error, unexpected CF_SYM_UNDEFINED" VyOS #1: # show protocols ospfv3 area 0.0.0.0 { interface wg3

3 4

eclist matching question
by Alexander Shikov 14 Jul '20

14 Jul '20

Hello! I have some strange behavior of list matching operator applied to eclist in v.1.6.3. Example filter function: function test (int peer_as) { if net = 109.68.40.0/21 && peer_as = 28773 then { print "DEBUG2: ", bgp_ext_community; } if (rt,65001,peer_as) ~ bgp_ext_community then { print "DEBUG: Adding one prepend: ", bgp_path.first; bgp_path.prepend(bgp_path.first); } } In log file I see bgp_ext_community attribute which includes (rt,65001,28773) community: Jun 15 15:37:52 <daemon.info> rs1 bird: DEBUG2: (eclist (rt, 65001, 13188) (rt, 65001, 28773) (ro, 31210, 25372)) But the later matching never succeeds, and I don't understand why. Please advice. Thanks! -- Alexander Shikov Technical Staff, Digital Telecom IX Tel.: +380 44 201 14 07 Mob.: +380 50 410 30 57 http://dtel-ix.net/

2 5

BIRD - RoA with aggregated prefixes - issue
by Javor Kliachev 14 Jul '20

14 Jul '20

Hello, We're using BIRD 1.6.4 as Route Server. Recently we have implemented ROA prefix validation but we have hit the issue with prefixes that are aggregated only. What do I mean: When the prefix is aggregate and has something like 1234 { 10, 20 } in AS_PATH in last asn, bgp_path.last value returns zero ( 0 ). As result of this we just discarding such prefixes. Our approach is the following: 1) We're using static RoA tables with prefixes for example: roa table r1234 { roa 10.10.10.0/24 max 32 as 1234; roa 10.10.11.0/24 max 32 as 1234; roa 10.10.12.0/24 max 32 as 1234; } 2) Then create a different function for each member like this and applying it on each protocol BGP as latest function: function AS1234_roa() { if roa_check(r1234, net, bgp_path.last) = ROA_INVALID then { print "ROA check failed: invalid prefix - ", net, " origin ASN ", bgp_path.last , " - AS-PATH", bgp_path , " via ", proto; return false; } if roa_check(r1234, net, bgp_path.last) = ROA_UNKNOWN then { print "ROA check failed: unallowed prefix - ", net, " origin ASN ", bgp_path.last , " - AS-PATH", bgp_path , " via ", proto; return false; } return true; } Could someone BIRD developer to suggest some solution for this issue? Thanks in advance! Best~ -- --- Javor Kliachev Senior Engineer IP Services office: +359 2 974 33 11 mobile: +359 885 98 84 95 [ http://www.neterra.net/ | www.neterra.net ] [ https://bg.linkedin.com/pub/javor-kliachev/11/b46/843 | ]

3 4

Migration from bird 1.6 to 2.0.
by Mike Neo 10 Jul '20

10 Jul '20

Hi, is it recommended to migrate from bird 1.6 to 2.0 if I use bird for an easy bgp router with 2-3 peers? What's the difference between 1.6 and 2.0? Best regards, Mike

3 2