- Bird-users - Mailing Lists

Bird v2, MPLS and LDP
by Maximilian Wilhelm 19 Sep '17

19 Sep '17

Hi folks, I've just seen in the NEWS for bird 2.0.0-pre1 that there are many MPLS related things coming. Kudos for that! Forgive me if this has been answered recently, I didn't find anything in my list archive: As of the available support for MPLS forwarding in the Linux kernel, I'm wondering if there is any rough roadmap the next releases? Are there any plans on integrating some LDP module (maybe also borrowing from the openldpd as FRR did? ;)) into bird? Thanks for your great work! Will someone of you folks at DENOG9? Best Max -- "Kunst ist Besessenheit. Du willst etwas unter Kontrolle bringen bevor es Dich kontrolliert." -- O

2 3

FreeBSD 11.1 OSPF multicast broken?
by Aragon Gouveia 19 Sep '17

19 Sep '17

Good day, There seems to be a multicast related bug with FreeBSD 11.1 and Bird 1.6.3. In summary: Two problems: * Ethernet interfaces configured as "broadcast" do not emit any OSPF packets, as observed by tcpdump. * Above interfaces receive OSPF packets from Linux neighbours (as seen by tcpdump), but Bird does not receive them (nothing logged in debug mode). Two workarounds: * Enabling "real broadcast" solves both problems above, but it must be enabled on all peers' configs so that everyone broadcasts. * Configuring interfaces as nbma on all peers also solves both problems. While Bird is running, there seem to be no OSPF multicast memberships reported by ifmcstat(8), eg. # ifmcstat -i vtnet1 -v vtnet1: inet 192.168.102.16 igmpv3 rv 2 qi 125 qri 10 uri 3 group 224.0.0.1 mode exclude mcast-macaddr 01:00:5e:00:00:01 link 52:54:00:85:d7:51 group 01:00:5e:00:00:01 I have also tried disabling firewalls, and have not seen any "Permission denied" socket errors logged by Bird in debug mode. I'd be delighted to help test/find a fix with my lab setup, but for now I'm a bit lost for digging into code and kernel APIs, which is where I suspect the problem lies. Any ideas where to start? Thanks, Aragon

3 9

BIRD export reject not working
by Marek Królikowski 19 Sep '17

19 Sep '17

Hello guys, I got strange problem with BGP i share data to Collector but i put to reject some bgp path.... This is how look configuration of BGP session of Collector: protocol bgp Collector { description "Collector"; local as MY-ASN; source address IPv61; neighbor IPV62 as SOMEONE-ASN; multihop; import none; export filter { if bgp_path.first = 64515 then reject; if bgp_path = 64515 then reject; if bgp_path.last = 64515 then reject; include "export.conf"; }; } But: root@pl-bgp:~# birdc6 show route export Collector all 2a0b:e206::/32 BIRD 1.6.3 ready. 2a0b:e206::/32 via 2a06:1287:330e:16::2 on bgp-fr [France 2017-09-18] * (100) [AS21430i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: MY-ASN 64515 20473 1299 21430 BGP.next_hop: 2a07:XXXX.XXXX.XXXX BGP.local_pref: 100 BGP.community: (64515,44) Like You see in BGP path there is 64515. How i export this to Collector if i put reject on export? What is wrong? I use BIRD 1.6.3. Best Regards Marek Krolikowski

4 4

Kernel dropped some netlink messages, will resync on next scan.
by Giuseppe Ravasio (LU) 13 Sep '17

13 Sep '17

Hi, I'm upgrading my core router (running an old bird 1.3.10) with a new Ubuntu 16.04 running 1.6.3 from the official PPA. Actually it'a a pretty simple configuration ISP3 ISP1 ISP2 | | | | | | NEWROUTER=====<iBGP>======OLDROUTER | | ipv4/21 ipv6/32 In the new router i find a lot of messages "Kernel dropped some netlink messages, will resync on next scan." for both IPv6 and IPv4 For IPv6 daemon this seems to happen only when i start the sessions and load the full table and so I think it could be acceptable. For IPv4 daemon this seems to happen *a lot* when I start the BGP session with the iBGP enabled (and also this I think could be acceptable), but the message gets printed sometimes (a single line every 5/10 minutes) when the BGP is running normally. I already tried to adjust the net.core mem kernel parameters with: net.core.rmem_max = 8388608 net.core.wmem_max = 8388608 net.core.optmem_max = 65536 net.core.rmem_default=1064960 net.core.wmem_default=1064960 But the messages are still there. Any hints? maybe the rmem_default is still too low? Thanks Giuseppe

3 6

documentation on emptying bgp_community
by Alexander Zubkov 13 Sep '17

13 Sep '17

Hello. I have found a couple of methods to make bgp_community empty: 1) bgp_community.empty 2) bgp_community = -empty- But neither of them are present in the documentation. Only example in bird.conf.example file. I think it will be good to have this documented.

2 1

reload configuration without changing protocol state
by Alexander Zubkov 12 Sep '17

12 Sep '17

Hello. I want to be able to reload bird's configuration so it will not change current protocol's state. For example if I run bird with some protocols and then disable one of them from the console and at finally I run "configure" of "configure soft" to reload a configuration (new filters for example) - I get that protocol restarted too: bird> show protocols name proto table state since info device1 Device master up 2017-09-09 static1 Static master up 18:07:29 bird> disable static1 static1: disabled bird> show protocols name proto table state since info device1 Device master up 2017-09-09 static1 Static master down 18:10:01 bird> configure soft Reading configuration from /etc/bird.conf Reconfigured bird> show protocols name proto table state since info device1 Device master up 2017-09-09 static1 Static master up 18:10:10 bird> But I want protocol to still be disabled in this case. Because I have disabled it manually and want it to be started manually. Is there a way of reloading configuration, keeping manual state of protocols? I have already tried documentation and Google, but with no luck. >

1 0

Re: BIRD IP tunnel problem
by Ondrej Zajicek 12 Sep '17

12 Sep '17

Hi BIRD does not handle well situations where there are overlapping IP ranges on interfaces. It just find a first iface that matches the next-hop. It is true that one would expect that in 'direct' mode, the iface is used also for the next hop, but even in that case BGP is two-level and BGP next hop is resolved through iface prefixes to find the next hop iface. You could try attached patch that will change the behavior that in 'direct' mode always use the iface over which the BGP session is configured. -- 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."

1 0

BIRD IP tunnel problem
by noc＠joey-network.de 12 Sep '17

12 Sep '17

Hi, I figured out a problem with the BIRD routing daemon. It is not possible to route BGP via IPIP tunnels. As you can see below the packets will be routed through the wrong device. Normally it should choose "ASIX" as right device, but it does not. root@cr01:/etc/bird# ip -6 route | grep 2a06:8187:fe48::/48 2a06:8187:fe48::/48 via 2a06:8187:fe01:100::2 dev vlan1000 proto bird src 2a06:8187:fe01::2 metric 1024 pref medium The interface configuration is the following: iface ASIX inet manual pre-up ip -6 tunnel add ASIX mode any local 2a07:59c6:d007::1 remote 2a00:1ca8:2a::730 ttl 255 pre-up ip link set dev ASIX mtu 1476 pre-up ip link set dev ASIX up pre-up ip address add 89.46.125.149/30 dev ASIX pre-up ip address add 2a06:8187:fe01:100::1/56 dev ASIX post-down ip -6 tunnel del ASIX root@cr01:/etc/bird# birdc6 -v 0001 BIRD 1.6.3 ready. bird> Maybe you can help me here? Kind regards. Joey

2 1

Bird redundant routers with same AS on same location and prefix delegation to lower router
by Gelderman, Richard 07 Sep '17

07 Sep '17

I am trying to achieve a redundant routing setup with two routers with the same ASN who do prefix delegation to a lower router. The setup is the following: AS1000 ISP1 ISP2 AS2000 | | | | AS10000 R1 ======== R2 AS10000 prefix /22 \ iBGP / prefix /22 \ / \ / \ / R3 AS65000 / \ / \ net A net B prefix /24 prefix /24 Does anybody has experience with this kind of setup? Regards, Richard

2 1

Monthly Reminders
by Jason Kopacko 01 Sep '17

01 Sep '17

Can we get the password field removed from the monthly emails? I find that highly bad juju emailing out passwords every month. Sent from my Verizon Wireless 4G LTE Droid Turbo 2.

1 0