- Bird-users - Mailing Lists

[PATCH] Multiple neighbor range protocols in BGP
by Liam Nattrass 17 Sep '19

17 Sep '19

Hey all, I was doing some work with dynamic neighbors in BGP, and found that I am not able to use multiple protocol definitions with neighbor ranges. Connections to the second and subsequent definitions are rejected. The BGP protocol for the subsequent instances depend on a lock, but due to the remote_addresses being the same (null) the subsequent protocols hang, waiting for the first protocol. Attached is a small patch which checks if the protocol is dynamic, and uses the remote_range prefix instead of a zero remote_address for the lock. Hope this helps. Thanks, Liam Nattrass

2 3

BIRD 2.0.6 and 1.6.8
by Ondrej Filip 15 Sep '19

15 Sep '19

Dear BIRD users, we just released two new versions - 1.6.8 and 2.0.6. Version 1.6.8 just includes important bugfixes. Version 2.0.6 o RAdv: Solicited unicast RAs o BGP: Optional Adj-RIB-Out o BGP: Extended optional parameters length o Filter: Sets and set expressions in path masks o Several important bugfixes Cheers! Ondrej

4 5

Advices for a weird multihop setup
by Alarig Le Lay 14 Sep '19

14 Sep '19

Hi, I have a weird setup with a transit. My next-hop is 2a03:94e0:feff:: (with a multihop session) which is reachable via 2a03:94e0:17ff::1 which is on-link on eno1. (your should already tear blood here) So, when I establish the session, the next-hop couldn’t be resolved and the route is marked as unreachable. e.g. 2a00:9600::/47 unreachable [bgp_terrahost_ipv6 12:32:10.864 from 2a03:94e0:feff::] * (100/-) [AS25605i] Type: BGP univ BGP.origin: IGP BGP.as_path: 203629 6939 3356 25605 BGP.next_hop: 2a03:94e0:feff:: BGP.local_pref: 100 So, for now I’m just exporting route with a next-hop different of 2a03:94e0:feff:: and I added a default route (ip -6 r a default via 2a03:94e0:17ff::1 dev eno1 onlink) to a least do the necessary routing on those. 40 protocol kernel kernel_ipv6 { 41 learn; 42 persist; 43 ipv6 { 44 import all; 45 export where bgp_next_hop != 2a03:94e0:feff::; 46 }; 47 } This seems to be a bit ugly for me, so I have to questions: * Is it possible to handle with on-link next-hop with bird? * I think not, so is there a less ugly work-around? 75 protocol static static_terrahost_ipv6 { 76 ipv6 { 77 import all; 78 }; 80 route 2a03:94e0:feff::/128 recursive 2a03:94e0:17ff::1; 81 route 2a03:94e0:17ff::1/128 via "eno1"; 82 } 257 protocol bgp bgp_terrahost_ipv6 from UPSTREAM_IPv6 { 258 description "terrahost"; 259 neighbor 2a03:94e0:feff:: as 203629; 260 multihop 3; 261 262 ipv6 { 263 import filter bgp_filter_terrahost_in_ipv6; 264 }; 265 } Regards, -- Alarig

2 2

Debian packages for BIRD 2
by Ondrej Zajicek 14 Sep '19

14 Sep '19

Hello Several days ago i received info that BIRD 2.0.4 is now packaged in Debian (unstable): https://packages.debian.org/sid/bird2 Thanks to Ondrej Sury and others. -- 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."

4 5

bird >=2.0.4 Stack underflow and most routes become filtered
by Дружок 1 13 Sep '19

13 Sep '19

Hi! .. /var/log/bird.log 2019-08-22 15:38:38.604 <ERR> filters, line 1023: Stack underflow 2019-08-22 18:02:27.157 <ERR> filters, line 511: Stack underflow 2019-08-22 18:02:30.616 <ERR> filters, line 0: Stack underflow.... ... function bgp_in(int peer_as; prefix set peer_nets) { if ! (avoid_martians()) then return false; if ! (net ~ peer_nets) then return false; if (bgp_path.first != peer_as ) then return false; # if (is_black()) then return true; # if net.len > 24 then return false; return true; } .. define DataClub_nets = [ 46.183.216.0/21{21,32} <http://46.183.216.0/21%7B21,32%7D>, 84.38.128.0/24{24,32} <http://84.38.128.0/24%7B24,32%7D>, 84.38.130.0/24{24,32} <http://84.38.130.0/24%7B24,32%7D>, 84.38.131.0/24{24,32} <http://84.38.131.0/24%7B24,32%7D>, 84.38.132.0/24{24,32} <http://84.38.132.0/24%7B24,32%7D>, 84.38.134.0/24{24,32} <http://84.38.134.0/24%7B24,32%7D>, 84.38.135.0/24{24,32} <http://84.38.135.0/24%7B24,32%7D>, 109.248.147.0/24{24,32} <http://109.248.147.0/24%7B24,32%7D>, 109.248.148.0/24{24,32} <http://109.248.148.0/24%7B24,32%7D>, 109.248.149.0/24{24,32} <http://109.248.149.0/24%7B24,32%7D>, 185.220.116.0/24{24,32} <http://185.220.116.0/24%7B24,32%7D>, 185.220.117.0/24{24,32} <http://185.220.117.0/24%7B24,32%7D>, 185.220.118.0/24{24,32} <http://185.220.118.0/24%7B24,32%7D>, 194.165.16.0/23{23,32} <http://194.165.16.0/23%7B23,32%7D> ]; .. protocol bgp DataClub from PEERS { description "DataClub/52048/as-dataclub"; neighbor 193.27.39.4 as 52048; ipv4 { import where bgp_in(52048, DataClub_nets); export where bgp_out(52048, 64004); }; } ... I redid the configuration something like ... ipv4 { import filter myfilter; .. ... then I issue test "show route filter myfilter" and get all nets. But in fact, all these routes were filtered. And, finally, only "accept" remained in the filter - this method unfiltered routes. Bug, feature? P.S.: Bird <= 2.0.3 is OK.

5 7

Filter ext_community?
by Chris Herdt 12 Sep '19

12 Sep '19

I have a number of routes that include non-standard bgp_ext_community entries, for example unknown 0x10b (VRF Route Import). Is there a way to filter bgp_ext_community for only route targets? I tried the following, but the ec datatype expects integers, not wildcards: show route filter { if (rt, *, *) ~ bgp_ext_community then accept; reject; } all I'm using bird 2.0.2. Thanks! -- Chris Herdt

2 1

ospf authentication status from bird cli?
by Kenth Eriksson 12 Sep '19

12 Sep '19

Is it possible to see ospf authentication status from bird cli? I have seen that authentication failures are logged to file, but it would be good to have it visible in cli as well. Thanks, Kenth

3 5

bfd or ospf problem?
by Dmitry Melekhov 12 Sep '19

12 Sep '19

Hello! I run two Centos 7 servers. They are connected over ipsec+gre, both run bird 2.0.5 ospf. I also have bfd enabled: bfd yes; in this interface and protocol bfd { interface "gre*" { interval 200 ms; }; }; Both sides had following messages in logs: Sep 12 08:18:23 ast-p245a bird: bfd1: Bad packet from 192.168.201.85 - unknown session id (1935426703) Sep 12 08:18:45 ast-ngdu2 bird: bfd1: Bad packet from 192.168.201.86 - unknown session id (913048279) And I had ospf error messages: Sep 8 03:31:04 ast-p245a bird: MyOSPF: Authentication failed for nbr 192.168.46.74 on gre2 - lower sequence number (rcv 1565416592, old 1566997226) Sep 8 03:31:04 ast-p245a bird: MyOSPF: Authentication failed for nbr 192.168.46.74 on gre2 - lower sequence number (rcv 1565416592, old 1566997226) Sep 8 03:31:04 ast-ngdu2 bird: MyOSPF: Authentication failed for nbr 192.168.26.2 on gre2 - lower sequence number (rcv 1565670907, old 1567089417) Sep 8 03:31:04 ast-ngdu2 bird: MyOSPF: Authentication failed for nbr 192.168.26.2 on gre2 - lower sequence number (rcv 1565670907, old 1567089417) One of servers got NetworkManager problem as result today Sep 8 03:31:21 ast-p245a NetworkManager[987]: <info> [1567899081.4736] platform-linux: netlink: read: too many netlink events. Need to resynchronize plat form cache There are many such messages in log And had connectivity problem, so I rebooted it and disabled bfd on both sides. Problem disappeared. May be there were problem on link, but I have two another servers on this link with identical config, but they run quagga. Don't sure this is bird problem, but may be somebody had the same problem before? Thank you!

1 0

Kernel protocol and IPv6 route issue with Bird 2.0
by Nigel Kukard 11 Sep '19

11 Sep '19

Hi guys, I'm having an issue where IPv6 routes from the kernel routing table don't appear to end up in my t_kernel6 table. Despite the next-hop being link-local and the interface being UP. IPv4 routes I have no problem with, working as expected. I must be doing something wrong on my side, if someone could kindly point me in the right direction. (the below is configuration I'm testing with, I have manually set the MAC addresses and IP's for testing purposes within a LXC container so I can test results across configuration changes) Here is the snippet of config with... protocol device { description "Device protocol"; scan time 10; }; protocol kernel kernel4 { description "Kernel protocol for IPv4"; metric 600; learn; ipv4 { table t_kernel4; export all; import all; }; }; protocol kernel kernel6 { description "Kernel protocol for IPv6"; metric 600; learn; ipv6 { table t_kernel6; export all; import all; }; }; Here is the error I'm getting in my logs and fc11::/64 is not being shown in my t _kernel6 table... 2019-09-11 11:40:36.653 <ERR> KRT: Received route fc11::/64 with strange next-hop fc10::2 The interface is however UP and fc10::2 is link local. Here is the output of 'ip link'... (eth1 is UP) 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth0@if3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000 link/ether 02:01:00:00:00:01 brd ff:ff:ff:ff:ff:ff link-netnsid 0 4: *eth1*@if5: <BROADCAST,MULTICAST,UP,*LOWER_UP*> mtu 1500 qdisc noqueue state *UP* mode DEFAULT group default qlen 1000 link/ether 02:01:00:00:00:02 brd ff:ff:ff:ff:ff:ff link-netnsid 0 Here is the output of 'ip -6 address'... 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 state UNKNOWN qlen 1 inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0@if3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 fc00::1/64 scope global valid_lft forever preferred_lft forever inet6 fe80::1:ff:fe00:1/64 scope link valid_lft forever preferred_lft forever 4: eth1@if5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 *fc10::1/64* scope global valid_lft forever preferred_lft forever inet6 fe80::1:ff:fe00:2/64 scope link valid_lft forever preferred_lft forever Here is the kernel routing table 'ip -6 route'... fc00::/64 dev eth0 proto kernel metric 256 pref medium fc10::/64 dev eth1 proto kernel metric 256 pref medium *fc11::/64 via fc10::2 dev eth1 metric 1024 pref medium* fe80::/64 dev eth0 proto kernel metric 256 pref medium fe80::/64 dev eth1 proto kernel metric 256 pref medium Kind Regards -N

2 2

Two similar kernel routes with different metrics on bird 1.6.4
by Andrew 11 Sep '19

11 Sep '19

Hi all. I have two default kernel routes with different metrics (for failover, when first gateway becomes unreachable - all traffic is switched to second gw). I want to pipe it to other kernel table to gracefully handle connections to second IP. Unfortunatelly, I can't force bird to learn route with higher metric - bird just ignores it. Option 'merge paths' does nothing in that case. Import filter doesn't help - it seems like route is dropped somewhere in protocol, before passing to filter. How can I force bird to learn route with higher metric?

2 1