- Bird-users - Mailing Lists

The function return true but it could not be export
by 智慧雨泽 08 Jun '21

08 Jun '21

Hi,The function i wrote correctly returns net. bird> show route 2602:feda:ab2::/48 where bgp_export_all() Table master6: 2602:feda:ab2::/48 unicast [LL_IX 02:19:31.236] * (100) [i] via 2602:feda:ab2::1 on ll-ix But it could be export. bird> show route export TOHU_IX for 2602:feda:ab2::/48 bird> I think that should be accept export filter { if bgp_export_all() then accept; # if (roa_check(r6, net, bgp_path.last_nonaggregated) ~ [ROA_INVALID, ROA_UNKNOWN]) then reject; # if is_bogon_asn() then reject; # if is_bogon_prefix() then reject; # accept; }; }; } Why is it not exported? The full configuration is. protocol bgp TOHU_IX { local as LOCAL_ASN; description "tohu ix IX-VPS BGP"; direct; source address 2602:feda:ab5::1; neighbor 2602:feda:ab6::1%'tuhu-ix' as 141011; ipv6 { next hop self; import filter { if is_bogon_asn() then reject; if is_bogon_prefix() then reject; if (roa_check(r6, net, bgp_path.last_nonaggregated) = ROA_VALID) then reject ; accept; }; export filter { if bgp_export_all() then accept; # if (roa_check(r6, net, bgp_path.last_nonaggregated) ~ [ROA_INVALID, ROA_UNKNOWN]) then reject; # if is_bogon_asn() then reject; # if is_bogon_prefix() then reject; # accept; }; }; } function bgp_export() { if net !~ OUR_PREFIXES then return false; if is_bogon_asn() then return false; if is_bogon_prefix() then return false; if net_len_too_long() then { # print "net ", net, " prefix too long"; return false; } if proto = "BGP_Prefix" then return true; if proto = "LL_IX" then{ if bgp_path.len = 0 then{ return true; } } if proto = "TOHU_IX" then{ if bgp_path.len = 0 then{ return true; } } return false; } function bgp_export_all() { if bgp_export() then { return true; } if source != RTS_BGP then{ return false; } if is_bogon_asn() then return false; if is_bogon_prefix() then return false; return true; }

3 4

Bird Interference with systemd-networkd
by Marvin Gaube 28 May '21

28 May '21

Hi, on one of my systems, I'm migrating from ifupdown2 to systemd-networkd. The scenario is an wireguard-overlay running OSPF and BGP using bird 2.0.7, and the underlay/public network interface being configured using DHCP and SLAAC. During the setup, I ran into problems with systemd-networkd, which changed the interface state into "Could not set NDisc route: Connection timed out" and loosing the Default WAN Route some time after bird installed the routes. This seems like a problem on a netlink-level etc. - likely on the networkd side, but I wonder if someone on this list had a similar problem. Bugreport and Discussion is ongoing over at systemd github: https://github.com/systemd/systemd/issues/19738 Thanks! Marvin

2 1

BIRD traffic
by Mike Marchal 26 May '21

26 May '21

Hi there, I use BIRD for my personal network and i'm trying to figure out if i can see a traffic consumed per IP somewhere, or create that output to php. Thanks, Best regards, Mike Marchal

3 2

Sticked routers
by Alexey 26 May '21

26 May '21

Hello, my bird is periodicy routes stick. Logs: -- Boot 168108ca9fc345ddaf52baeeaed51516 -- May 26 03:48:51 router systemd[1]: Starting BIRD routing daemon... May 26 03:48:51 router systemd[1]: Started BIRD routing daemon. May 26 03:48:51 router bird[542]: Started May 26 03:48:51 router bird[542]: Next hop address 198.18.2.1 resolvable through recursive route for 0.0.0.0/0 May 26 03:48:51 router bird[542]: Next hop address 2a01:*:*:1::1 resolvable through recursive route for ::/0 May 26 03:48:51 router bird[542]: Next hop address 2a01:*:*:: resolvable through recursive route for ::/0 May 26 03:48:51 router bird[542]: Netlink: Invalid argument May 26 03:48:51 router bird[542]: Netlink: Invalid argument May 26 03:48:52 router bird[542]: Netlink: Invalid argument May 26 03:48:52 router bird[542]: Netlink: Invalid argument May 26 03:48:52 router bird[542]: Netlink: Invalid argument May 26 03:48:52 router bird[542]: ... May 26 03:48:53 router bird[542]: Netlink: Invalid argument May 26 03:48:57 router bird[542]: bfd1: Bad packet from 169.254.100.6 - unknown session id (3861494737) May 26 03:48:57 router bird[542]: bfd1: Bad packet from 169.254.100.6 - unknown session id (3861494737) May 26 03:48:57 router bird[542]: bfd1: Bad packet from 169.254.100.6 - unknown session id (3861494737) May 26 03:48:57 router bird[542]: bfd1: Bad packet from 169.254.100.6 - unknown session id (3861494737) May 26 03:48:57 router bird[542]: bfd1: Bad packet from 169.254.100.6 - unknown session id (3861494737) May 26 03:48:58 router bird[542]: bfd1: Bad packet from 169.254.101.4 - unknown session id (2963707952) May 26 03:48:58 router bird[542]: bfd1: Bad packet from 169.254.101.4 - unknown session id (2963707952) May 26 03:48:58 router bird[542]: bfd1: Bad packet from 169.254.101.4 - unknown session id (2963707952) May 26 03:48:58 router bird[542]: bfd1: Bad packet from 169.254.101.4 - unknown session id (2963707952) May 26 03:48:58 router bird[542]: bfd1: Bad packet from 169.254.101.4 - unknown session id (2963707952) May 26 03:56:10 router bird[542]: Next hop address 2002:c612:78ff:1::6 resolvable through recursive route for ::/0 May 26 07:23:28 router bird[542]: bfd1: Bad packet from 169.254.100.4 - unknown session id (2189589031) May 26 07:23:28 router bird[542]: bfd1: Bad packet from 169.254.100.4 - unknown session id (2189589031) May 26 07:23:28 router bird[542]: bfd1: Bad packet from 169.254.100.4 - unknown session id (2189589031) May 26 07:23:28 router bird[542]: bfd1: Bad packet from 169.254.100.4 - unknown session id (2189589031) May 26 07:23:28 router bird[542]: bfd1: Bad packet from 169.254.100.4 - unknown session id (2189589031) May 26 07:23:28 router bird[542]: bfd1: Bad packet from 169.254.100.4 - unknown session id (2189589031) May 26 07:25:18 router bird[542]: AS4200220007_2: Error: Hold timer expired May 26 07:27:16 router bird[542]: bfd1: Bad packet from 169.254.101.4 - invalid flags (240) Routes received from AS4200220007_2 is not removed from kernel tables. Kernel: 5.12.5-arch1-1 Destribution: Arch Linux

1 0

[PATCH] Add mips Debian 9 container
by Skyler Mäntysaari 24 May '21

24 May '21

I hope that this will help with the mips package building. The mips qemu emulation should be enabled with the: "docker --rm aptman/qus -s -- -p mips" so that it it would even work at all. --- .gitlab-ci.yml | 10 ++++++++++ misc/docker/debian-9-mips/Dockerfile | 26 ++++++++++++++++++++++++++ 2 files changed, 36 insertions(+) create mode 100644 misc/docker/debian-9-mips/Dockerfile diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 3642b671..f6a5bb39 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -17,6 +17,7 @@ stages: allow_failure: true script: - $DOCKER_CMD login -u gitlab-ci-token -p $CI_BUILD_TOKEN registry.labs.nic.cz + - $DOCKER_CMD --rm --privileged aptman/qus -s -- -p mips # Make sure we refresh the base image if it updates (eg. security updates, etc) # If we do just the build, cache is always reused and the freshness of the # base image is never checked. However, pull always asks and updates the @@ -52,6 +53,11 @@ docker_debian-9-i386: IMG_NAME: "debian-9-i386" <<: *docker_build +docker_debian-9-mips: + variables: + IMG_NAME: "debian-9-mips" + <<: *docker_build + docker_debian-10-amd64: variables: IMG_NAME: "debian-10-amd64" @@ -248,6 +254,10 @@ build-debian-9-i386: <<: *build-linux image: registry.labs.nic.cz/labs/bird:debian-9-i386 +build-debian-9-mips: + <<: *build-linux + image: registry.labs.nic.cz/labs/bird:debian-9-mips + build-debian-10-amd64: <<: *build-linux image: registry.labs.nic.cz/labs/bird:debian-10-amd64 diff --git a/misc/docker/debian-9-mips/Dockerfile b/misc/docker/debian-9-mips/Dockerfile new file mode 100644 index 00000000..1600011b --- /dev/null +++ b/misc/docker/debian-9-mips/Dockerfile @@ -0,0 +1,26 @@ +FROM multiarch/debian-debootstrap:mips-stretch-slim +ENV DEBIAN_FRONTEND noninteractive +RUN sed -i 's/deb.debian.org/ftp.cz.debian.org/' /etc/apt/sources.list +RUN apt-get -y update +RUN apt-get -y upgrade +RUN apt-get -y --no-install-recommends install \ + build-essential \ + flex \ + bison \ + autoconf \ + ncurses-dev \ + libreadline-dev \ + libssh-gcrypt-dev \ + linuxdoc-tools-latex \ + texlive-latex-extra \ + opensp \ + docbook-xsl \ + xsltproc +RUN apt-get -y --no-install-recommends install \ + git \ + dpkg-dev \ + debhelper \ + quilt \ + python3 \ + python3-pip \ + python3-setuptools -- 2.26.3

1 0

Re: BIRD packages in Gitlab CI
by Skyler Mäntysaari 21 May '21

21 May '21

Unfortunately, doesn't change the fact that Bird 2.0.8 is not available on mips. The Gitlab CI should be extended to include that too. P.S Do not email me off-list please. On 21/05/2021 8.05, Kees Meijs | Nefos wrote: > On https://packages.debian.org/buster-backports/bird2 you can find all > flavours. > > K. > > On 21-05-2021 06:22, Skyler Mäntysaari wrote: >> Still x86_64 and i386 platforms only. :( >> >> On 20/05/2021 17.18, Ondrej Zajicek wrote: >>> Hello >>> >>> Finally we managed to integrate apkg into Gitlab CI and now we not only >>> have apkg metafiles in the master branch (so anybody can use apkg to >>> build deb or rpm packages for their distribution), but we also have >>> automatic build of packages in Gitlab CI for most mainstream >>> distributions (Debian, Ubuntu, Fedora, CentOS, OpenSUSE), which can be >>> downloaded as artifacts [*] for each commit. Thanks to Jakub Ruzicka and >>> Matous Holinka for some work on this. >>> >>> It is not as comfortable as fully automated and signed public >>> repository, >>> but still better than compile it yourself approach :-). >>> >>> [*] Right side drop menus on https://gitlab.nic.cz/labs/bird/-/pipelines >>>

4 3

config.Y grammar for a new protocol
by Vinod Padmanabhan 21 May '21

21 May '21

Hi, I would like to know how to add a config.Y file for any new protocol in Bird2.0. For example, if my config file looks like: ipv4 table xyz; protocol new { ipv4 { table xyz; import all; export where source = RTS_BGP; }; } How will the corresponding grammar look like? Thanks, Vinod

2 1

BIRD packages in Gitlab CI
by Ondrej Zajicek 21 May '21

21 May '21

Hello Finally we managed to integrate apkg into Gitlab CI and now we not only have apkg metafiles in the master branch (so anybody can use apkg to build deb or rpm packages for their distribution), but we also have automatic build of packages in Gitlab CI for most mainstream distributions (Debian, Ubuntu, Fedora, CentOS, OpenSUSE), which can be downloaded as artifacts [*] for each commit. Thanks to Jakub Ruzicka and Matous Holinka for some work on this. It is not as comfortable as fully automated and signed public repository, but still better than compile it yourself approach :-). [*] Right side drop menus on https://gitlab.nic.cz/labs/bird/-/pipelines -- Elen sila lumenn' omentielvo Ondrej 'Santiago' Zajicek (email: santiago(a)crfreenet.org) OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net) "To err is human -- to blame it on a computer is even more so."

2 1

Exact Definition of numbers-match bitmask-match and fragmentation-type
by Matt Corallo 18 May '21

18 May '21

The match classifiers for flowspec (numbers-match bitmask-match and fragmentation-type) don't appear to be exactly specified in the documentation anywhere. eg 1) Based on the grouping in the examples, I'd think && and || are parsed first, followed by , and .. ie 1 && 2,3 means 1 AND (two OR three) not (1 AND 2) or 3. Same would apply for fragmentation-type, which makes the spacing in the flow4 example somewhat confusing, though obviously it doesn't change the logic given its all ORs. 2) What are the full list of possible operators? From the examples there is at least <, <=, >, >=, &&, ||, ",", .., and !. 3) I can make a pretty good guess as to what each operator means, but it doesn't seem to be written down anywhere. Separately, the documentation seems to indicate dscp applies for both IPv4 flowspec as well as IPv6 flowspec as-is, however this is somewhat confusing - is DSCP intended to match on v4 ECN bits as well, or is the expected length of DSCP simply 8 bits instead for v6 and matched against the traffic class? Finally, and I suppose this is an RFC question not a BIRD question, how does the fragment field interact with the next header field in v6 - I assume a router is expected to parse a fragment header before checking next header, irrespective of the fragment bits (unless they require all bits unset)? Thanks, Matt

4 19

[PATCH] Filter: Add MPLS label route attribute
by Trisha Biswas 17 May '21

17 May '21

Hi all, While BIRD 2.x has underlying support for MPLS (applying labels to static routes and syncing with kernel), there is no option to set labels via configuration. The attached patch adds support to set or read labels using filters. Currently this supports the addition of one label per route, and can be extended to stacked labels in future. Please let me know if you have any questions. Thank you, Trisha -- *Trisha Biswas* | Sr. Software Engineer, Network Systems fastly.com | @fastly <https://twitter.com/fastly> | LinkedIn <http://www.linkedin.com/company/fastly>

2 3