July 2021 - Bird-users - Mailing Lists

The internal BGP route was successfully exported but could not be imported
by 智慧雨泽 28 Jul '21

28 Jul '21

Hello everyone,I use a pure BGP network to form internal routes (because I want to keep the as-path), and the ASN of each router is the same. The vserver of LL_IX successfully export the route bird> show route for 2602:feda:ab2::/48 export HOME_1_ARCH Table master6: 2602:feda:ab2::/48 unreachable [BGP_Prefix 11:52:02.569] (200) But the server at my home could not import it,even i change to import all; bird> show route for 2602:feda:ab2::/48 protocol LL_IX bird> show route for 2602:feda:ab2:: protocol LL_IX bird> This is the code on home server template bgp tpl_ibgp { local as LOCAL_ASN; rr client; direct; ipv6 { next hop self; import all; # import filter { # if is_bogon_asn() then { # print "is bogon asn", net, " for ASN ", bgp_path.last; # reject; # } # if is_bogon_prefix() then { # print "is bogon prefix", net, " for ASN ", bgp_path.last; # reject; # } # if my_opt_prefix() then bgp_path.prepend(LOCAL_ASN); # accept; # }; export filter { if my_opt_prefix() then bgp_path.prepend(LOCAL_ASN); if bgp_export_all() then accept; reject; }; }; } And the log Jul 28 12:16:38 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:38 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:38 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:38 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:40 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:40 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:43 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:43 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:43 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:44 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:45 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:45 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:45 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:47 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:47 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:47 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:47 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:48 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. Jul 28 12:16:48 archlinux bird[71587]: Kernel dropped some netlink messages, will resync on next scan. And the vserver at ll_ix.and log on this server is just about rpki. template bgp tpl_ibgp { local as LOCAL_ASN; rr client; direct; ipv6 { next hop self; # import all; import filter { if is_bogon_asn() then { print "is bogon asn", net, " for ASN ", bgp_path.last; reject; } if is_bogon_prefix() then { print "is bogon prefix", net, " for ASN ", bgp_path.last; reject; } if my_opt_prefix() then bgp_path.prepend(LOCAL_ASN); accept; }; export filter { if my_opt_prefix() then bgp_path.prepend(LOCAL_ASN); if bgp_export_all() then accept; reject; }; }; } Under what circumstances BGP import will only be "ignored? "I think this is the problem. Could soamone help me ? thanks.

2 1

changing BGP default bgp_local_pref resets protocol
by Ross Tajvar 26 Jul '21

26 Jul '21

Hi, When I change the "default bgp_local_pref" value for a bgp protocol, it causes the session to drop and re-establish. Here is an example config: protocol bgp bgp_transit_companyname_v6 { > ipv6 { > import table; > import keep filtered; > import filter bgp_in_transit_companyname; > export filter bgp_out_transit_companyname; > }; > local as MY_AS; > neighbor 2001:db8::1 as 64496; > default bgp_local_pref 45; > } Is this supposed to happen? Is there a way to get it not to happen? I would expect this change to simply change the local preference on all of the routes I have received, which would be much less disruptive. Thanks, Ross

2 2

Filter babel routes on a per-neighbor or per-interface basis
by Fabian Bläse 26 Jul '21

26 Jul '21

Hi, is there any possibility to filter incoming babel routes on a per-neighbor or per-interface basis? As far as I can tell, the babel implementation in bird has its own route selection algorithm and only sends a single route to the bird table. This seems to make it impossible to add additional route filters (for example: adding extra penalty to specific routes received on a specific babel interface), because with regular filters only the already pre-selected route in the bird table is changed. Also, this makes it impossible to use ECMP with babel. Would it be possible to send all feasible babel routes to the bird table? Best regards, Fabian

3 3

BIRD - iBGP between RS over IPv6 lead to crash (segfault)
by Javor Kliachev 26 Jul '21

26 Jul '21

Hello, We're using BIRD 1.6.4 as Route Server. Recently we have decided to establish an internal BGP session between both RS over IPv6 But when we added our config in BIRD then it has crashed immediately with the following error: [35878512.553496] bird6[2545]: segfault at 0 ip 000055f60242c331 sp 00007ffd43689460 error 4 in bird6[55f60240a000+86000] At same time we have several working iBGP session between RS <> Routers ( not RS ) from a long time and everything works fine with them. Here is the example config that crashed the system. Note: The original ASN has been replaced with 1234 in the example. table T1234 protocol pipe P1234 from iBGP_PIPES { peer table T1234; export all; } protocol bgp R100 { description "iBGP_RS1"; source address 2001:67c:29f0::A:1:234:200; neighbor 2001:67c:29f0::A:1:234:100 as 1234; local as myas; prefer older on; missing lladdr ignore; import none; export none; passive off; table T1234; add paths on; direct; } Another more strange: Initially before the crash we have added the following config below and RS was stable but session was not able to establish. We have made a capture of the traffic during the attempt of session establishment and seems TCP handshake works but then not procced futher. Here is the config: template bgp iBGP { local as myas; import all; export all; import limit 21000; passive on; prefer older on; missing lladdr ignore; } protocol bgp R1234x100 from iBGP { description "iBGP_RS1"; neighbor 2001:67c:29f0::A:1:234:100 as 1234; import all; export all; passive off; table T1234; # add paths on; direct; import limit 4000000; } Here is the logs from both side: Jul 7 11:52:09 rs1 bird6: R1234x4: Connecting to 2001:67c:29f0:0:a:1:234:4 from local address 2001:67c:29f0:0:a:1:234:100 2021-07-07 11:53:56 <TRACE> R1234x100: Connecting to 2001:67c:29f0:0:a:1:1234:100 from local address 2001:67c:29f0:0:a:1:234:4 Upper logs messages are repeating several times but iBGP was not establish.No any other logs persist in the logs message regarding this session establishment attempt. Could someone BIRD developer to suggest some solution for fixing this issue? We highly appreciate any kind of support. I had doubt that "add paths on" command would leading to this crashes but I commented it and after reloading the system crashed again. At same time this setup iBGP between the same RS over IPv4 works fine. 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 | ]

2 4

WG: Problems configuring multihop BGP setup
by Rainer Kulow 23 Jul '21

23 Jul '21

I'am sorry, the ascii graphic was somehow wrapped a bit strangely. Here is a better version of it (now using the correct AS-Numbers): +-----------------+ | AS 60007 | |10.70.0.1/16| | | +-------+--------+ | 172.16.18.1/30 | | 172.16.18.2/30 +-------+--------+ | AS 60002 | |10.20.0.1/16| | | +-------+--------+ | 172.16.96.1/30 | | 172.16.96.2/30 +-------+--------+ | AS 60005 | |10.50.0.1/16| | | +-----------------+ On 22-07-2021 15:43, Rainer Kulow wrote: >Hello bird community, > >I am trying to establish complete connectivity using BGP in the following >setup containing three AS with direct links: > > 172.16.18.1/30 172.16.18.2/30 172.16.96.1/30 >172.16.96.2/30 >+--------------+ +--------------+ >+--------------+ >| AS 60007 +---------------------------+ AS 60002 >+---------------------------+ AS 60005 | >+--------------+ +--------------+ >+--------------+ >10.70.0.1/16 10.20.0.1/16 >10.50.0.1/16 > >The goal is that every address from one of the /16 networks can reach all >addresses in the other /16 networks. > >My current setup (here for the middle AS, the others are configured >analogously): >AS 60002 has three network cards, one with 172.16.18.2, one with 172.16.96.1 >and one 10.20.0.1. > >log syslog all; >router id 10.20.0.1; > ># interface scan every 10s >protocol device { > scan time 10; >} > ># no automaic generated direct routes to all network interfaces >protocol direct { > disabled; >} > ># synchronize BIRD routing tables with the OS kernel >protocol kernel { > ipv4 { > import all; > export all; > }; > merge paths on; >} > ># static IPv4 routes >protocol static static1{ > ipv4; > route 10.20.0.0/16 via 10.20.0.1; >} > ># BGP peers >template bgp bgp_peer{ > local as 60002; > multihop; > > ipv4 { > import filter { > if proto = "static1" then reject; > accept; > }; > export all; > next hop self; > }; > >} > >protocol bgp a_b from bgp_peer { > description "BGP a-b"; > neighbor 172.16.18.1 as 60007; >} > >protocol bgp b_c from bgp_peer { > description "BGP b-c"; > neighbor 172.16.96.2 as 60005; >} > > >If I delete "multihop" and "next hop self" from the template, I get all >routes, but can only ever ping the closest AS. >With "multihop" I also get all routes, but the routes to other AS are marked >as unreachable. >I would be glad if someone could tell me how to realize this multihop >scenario. It would also be interesting that the solution scales, so that I >can hang, for example, more AS in the chain above. > >Thanks in advance, >Best regards, >Rainer

3 5

RE: Problems configuring multihop BGP setup
by Rainer Kulow 22 Jul '21

22 Jul '21

I'am sorry, the ascii graphic was somehow wrapped a bit strangely. Here is a better version of it: +-----------------+ | AS 64517 | |10.70.0.1/16| | | +-------+--------+ | 172.16.18.1/30 | | 172.16.18.2/30 +-------+--------+ | AS 64512 | |10.20.0.1/16| | | +-------+--------+ | 172.16.96.1/30 | | 172.16.96.2/30 +-------+--------+ | AS 64515 | |10.50.0.1/16| | | +-----------------+ On 22-07-2021 15:43, Rainer Kulow wrote: >Hello bird community, > >I am trying to establish complete connectivity using BGP in the following >setup containing three AS with direct links: > > 172.16.18.1/30 172.16.18.2/30 172.16.96.1/30 >172.16.96.2/30 >+--------------+ +--------------+ >+--------------+ >| AS 60007 +---------------------------+ AS 60002 >+---------------------------+ AS 60005 | >+--------------+ +--------------+ >+--------------+ >10.70.0.1/16 10.20.0.1/16 >10.50.0.1/16 > >The goal is that every address from one of the /16 networks can reach all >addresses in the other /16 networks. > >My current setup (here for the middle AS, the others are configured >analogously): >AS 60002 has three network cards, one with 172.16.18.2, one with 172.16.96.1 >and one 10.20.0.1. > >log syslog all; >router id 10.20.0.1; > ># interface scan every 10s >protocol device { > scan time 10; >} > ># no automaic generated direct routes to all network interfaces >protocol direct { > disabled; >} > ># synchronize BIRD routing tables with the OS kernel >protocol kernel { > ipv4 { > import all; > export all; > }; > merge paths on; >} > ># static IPv4 routes >protocol static static1{ > ipv4; > route 10.20.0.0/16 via 10.20.0.1; >} > ># BGP peers >template bgp bgp_peer{ > local as 60002; > multihop; > > ipv4 { > import filter { > if proto = "static1" then reject; > accept; > }; > export all; > next hop self; > }; > >} > >protocol bgp a_b from bgp_peer { > description "BGP a-b"; > neighbor 172.16.18.1 as 60007; >} > >protocol bgp b_c from bgp_peer { > description "BGP b-c"; > neighbor 172.16.96.2 as 60005; >} > > >If I delete "multihop" and "next hop self" from the template, I get all >routes, but can only ever ping the closest AS. >With "multihop" I also get all routes, but the routes to other AS are marked >as unreachable. >I would be glad if someone could tell me how to realize this multihop >scenario. It would also be interesting that the solution scales, so that I >can hang, for example, more AS in the chain above. > >Thanks in advance, >Best regards, >Rainer

1 0

Problems configuring multihop BGP setup
by Rainer Kulow 22 Jul '21

22 Jul '21

Hello bird community, I am trying to establish complete connectivity using BGP in the following setup containing three AS with direct links: 172.16.18.1/30 172.16.18.2/30 172.16.96.1/30 172.16.96.2/30 +--------------+ +--------------+ +--------------+ | AS 60007 +---------------------------+ AS 60002 +---------------------------+ AS 60005 | +--------------+ +--------------+ +--------------+ 10.70.0.1/16 10.20.0.1/16 10.50.0.1/16 The goal is that every address from one of the /16 networks can reach all addresses in the other /16 networks. My current setup (here for the middle AS, the others are configured analogously): AS 60002 has three network cards, one with 172.16.18.2, one with 172.16.96.1 and one 10.20.0.1. log syslog all; router id 10.20.0.1; # interface scan every 10s protocol device { scan time 10; } # no automaic generated direct routes to all network interfaces protocol direct { disabled; } # synchronize BIRD routing tables with the OS kernel protocol kernel { ipv4 { import all; export all; }; merge paths on; } # static IPv4 routes protocol static static1{ ipv4; route 10.20.0.0/16 via 10.20.0.1; } # BGP peers template bgp bgp_peer{ local as 60002; multihop; ipv4 { import filter { if proto = "static1" then reject; accept; }; export all; next hop self; }; } protocol bgp a_b from bgp_peer { description "BGP a-b"; neighbor 172.16.18.1 as 60007; } protocol bgp b_c from bgp_peer { description "BGP b-c"; neighbor 172.16.96.2 as 60005; } If I delete "multihop" and "next hop self" from the template, I get all routes, but can only ever ping the closest AS. With "multihop" I also get all routes, but the routes to other AS are marked as unreachable. I would be glad if someone could tell me how to realize this multihop scenario. It would also be interesting that the solution scales, so that I can hang, for example, more AS in the chain above. Thanks in advance, Best regards, Rainer

1 0

Log and Filter Question
by pr0j3ctx 13 Jul '21

13 Jul '21

Hello. I have two questions about bird2. 1. I've seen that you can use "debug channels all;" can turn on the channel log. Unfortunately I couldn't do that. Is this right? If I do so, I get "/etc/bird/bird.conf:22:7 syntax error, unexpected CF_SYM_UNDEFINED, expecting PROTOCOLS or COMMANDS or LATENCY" 2. I built various filters. I can also see my print message in the log. Can I also have this message displayed when I call up "birdc show route filtered". So which filter took effect? Or do I need another command for this? Best wishes pr0

2 3

Multiple router IDs
by William 13 Jul '21

13 Jul '21

Hi all, I'm wondering if there's a way to present a different router ID to different BGP (or other) peers? I can't see a way of doing it unless the "router id from <interface mask>" provides the solution but it's still a global option. Quagga had that function as part of its BGP multiple-instance mode, and for what I'm investigating, would be useful tool. Regards, William

2 1

Failure to initialize kernel protocol on Bird 1.6.8
by Mathew Heard 11 Jul '21

11 Jul '21

Hi, I've [suspect] discovered a bug in Bird 1.6.8 where different results are seen when "birdc configure" vs restarting the daemon. In this case "birdc configure" fails to initialize the protocol leaving the protocol in the "start" state permanently. NOTE: the protocol can be repaired on an existing BGP instance simply by changing the name and calling configure a second time. This appears to force a refresh of the protocol and gets it into the correct "up" state. Starting the daemon fresh with the configuration results in correct behaviour. To replicate add the following configuration to your configuration file: protocol pipe { table t_blah; peer table master; export filter no_default; import filter peer_table; } protocol static { table t_blah; route 0.0.0.0/0 recursive 10.0.0.1; } protocol kernel k_blah { table t_blah; kernel table 1234; scan time 2; persist; export filter conditional_table; import none; } The build of bird I am using is the official Debian release. Regards, Mathew

1 2