- Bird-users - Mailing Lists

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

PATCH: backport of graceful restart to 1.6.8
by Mathew Heard 11 Jul '21

11 Jul '21

Attached is a backport of graceful restart to 1.6.8 (legacy branch). I am aware that this branch is no longer getting feature updates. However I would request this patch inclusion on the basis it's valuable for anyone looking to move from 1.x to 2.x. Regards, Mathew

1 0

BIRD 2.0 Syntax issue with defined()
by Elmar K. Bins 10 Jul '21

10 Jul '21

Hello fellows, I have my bird config failing on something as trivial as if defined (some_variable_never_defined) then ... ; with "syntax error, unexpected CF_SYM_UNDEFINED" on the "some_...". Am I wrongly assuming that defined() would check for the variable to exist within BIRD's global scope (e.g., appear in the namespace)? How, if not this way, can I check for a variable to have been defined prior (and, if it does, do certain things, of course)? Elmar. PS: The first example on the filters page makes me believe that defined() works in exactly this way, but I might be mistaken.

3 4

direct routes and SLAAC addresses
by George Ross 09 Jul '21

09 Jul '21

Many thanks for answering my stubnet question a couple of weeks or so back. I have a supplementary direct-protocol question now... We have a Linux router running BIRD 2.0.8. It has some interfaces on which we configure IPv6 addresses statically. We turn off auto-configuration, but, due to a race condition which it's going to be hard to fix, when we reboot the machine we can end up with SLAAC addresses as well as the static IPv6 addresses. We start BIRD, and the direct protocol imports the subnets, so we can then export them to BGP. The problem seems to come when the kernel ages out the SLAAC addresses. It looks as though when a SLAAC address goes, the direct protocol notices and stops importing the corresponding subnet. Unfortunately, it looks as though it doesn't realise that there's still a static IPv6 address for the subnet, and that it should therefore continue to import it. Does that sound right, or am I missing some bit of configuration somewhere? (As a hack workaround we now delete all of those unwanted SLAAC addresses before we start BIRD.) Thanks again for your help, and for BIRD. It's been very useful! -- George D M Ross MSc PhD CEng MBCS CITP University of Edinburgh, School of Informatics, Appleton Tower, 11 Crichton Street, Edinburgh, Scotland, EH8 9LE Mail: gdmr(a)inf.ed.ac.uk Voice: 0131 650 5147 PGP: 1024D/AD758CC5 B91E D430 1E0D 5883 EF6A 426C B676 5C2B AD75 8CC5 The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

3 6