- Bird-users - Mailing Lists

Weird behavior under FreeBSD
by Jiří Kubíček 17 Aug '10

17 Aug '10

Hello, after few days of successful running bird under FreeBSD 8.1 it becomes weird. Load of the machine jumps to 2.00 and birdc respond's very slowly. It takes minutes to "display protocols" and it becomes worse and worse. Tried to upgrade from 1.2.3 to 1.2.4 and the same thing hapened today. Killing bird takes also a long time. After restarting the process the same behaviour can be observed. I have tried to disable protocols and found out that disabling kernel1 and restarting bird helped and birdc responded smoothly. After re-enabling kernel1 I have observed in log file that exporting to kernel1 is extremely slow (about 5 prefixes per second) and exporting 320k prefixes takes for ever (few seconds under normal conditions). The only workaround I have is rebooting the machine. I currenty run two bird routers peering each other and the same thing was observed on both of them today. One of them runs also bird6 and there was no problem observed. Is anyone running bird on FreeBSD without problems? Can someone give me any hints what do observe/debug if it happens again? Thank you Jiri KUBICEK -- KRAXNET s.r.o. - www.kraxnet.cz Kamenicka 26, 170 00 Praha 7 ---

2 2

mask problem
by Jaroslav Jirásek 17 Aug '10

17 Aug '10

Hi, I have pc router with bird and mikrotik router, ospf routing is configured and working. On mikrotik router I have in log message with every received hello packet "mismatch in network mask - mine 255.255.255.252, remote 255.255.255.248" This messages appears when I have on pc router two IP adress on one interface - one is public 77.x.x.x/30, second is privat 172.16.0.1/29 .When I remove privat ip adress, then messages in mikrotik routers disappears. there is a part of bird.conf interface "vlan10" { cost 10; priority 1; authentication none; type nonbroadcast; strict nonbroadcast yes; neighbors { 77.x.x.x; }; poll 10; retransmit 5; transmit delay 1; hello 10; dead 40; wait 40; }; Thank you for hint

2 1

"ip sla monitor" with bird ?
by Hervé COMMOWICK 13 Aug '10

13 Aug '10

Hello, I know this feature is highly cisco related, but i want to know if it can be planned for a future release or not ? For those who maybe doesn't know this kind of feature, i want to condition bgp advertisement or kernel route sync with an external check. By the way, i'm interested if someone have done similar mecanism with opensource software. Thanks ! -- Hervé COMMOWICK, EXOSEC (http://www.exosec.fr/) ZAC des Metz - 3 Rue du petit robinson - 78350 JOUY EN JOSAS Tel: +33 1 30 67 60 65 - Fax: +33 1 75 43 40 70 mailto:hcommowick@exosec.fr

2 4

Re: bird not listening on "odd" interfaces?
by Ondrej Zajicek 13 Aug '10

13 Aug '10

On Thu, Aug 12, 2010 at 09:43:46AM -0700, Ask Bj?rn Hansen wrote: > > On Aug 12, 2010, at 9:42, Ondrej Zajicek wrote: > > >> Yes, certainly - config is at http://tmp.askask.com/2010/08/buh1-bird.conf - show interfaces below. > > > > The problem is that the peer 80.96.148.1 is not directly reachable by any > > iface. If it is the other end of venet0, you should set IP addresses > > accordingly. > > Yeah, I think OpenVZ isn't exposing how it actually is working > (because the PtP interface makes it work somehow). But even in that case kernel should know where to send packets for 80.96.148.1 . Perhaps there is some route to that address? What is your routing table? What is a traceroute to 80.96.148.1 from that host? If there is more hops you should set multihop option. > However: Shouldn't bird at least listen to port 179 anyway? Then I > could just run it in passive mode. If BIRD didn't found appropriate iface, it didn't started the BGP protocol. Even if it would listen to port 179 (for example, if there is another proper BGP protocol), it wouldn't accept session from that IP because it wouldn't know the iface. -- 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

Re: bird not listening on "odd" interfaces?
by Ondrej Zajicek 12 Aug '10

12 Aug '10

On Thu, Aug 12, 2010 at 07:36:10AM -0700, Ask Bj?rn Hansen wrote: > > On Aug 12, 2010, at 1:42, Ondrej Zajicek wrote: > > > Could you send me your config file and an output of 'show interfaces' command? > > Yes, certainly - config is at http://tmp.askask.com/2010/08/buh1-bird.conf - show interfaces below. The problem is that the peer 80.96.148.1 is not directly reachable by any iface. If it is the other end of venet0, you should set IP addresses accordingly. -- 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 not listening on "odd" interfaces?
by Ask Bjørn Hansen 12 Aug '10

12 Aug '10

Hi everyone, First of all - thanks for such a nice package of software. The edges are a little rough when getting started; but the configuration file format is very nice and until this everything has "just worked". I'm using bird to announce an AS and a /24 for a anycast network. On to my question: One of the boxes I'm running bird on is an OpenVZ virtualized linux box with an unusual network configuration. It hasn't given any trouble for the regular DNS server duties, but I can't seem to get bird to play nicely. The networks I want to announce are on a dummy0 interface. 'venet0' is 127.0.0.1; 'venet0:0' is a point to point interface with the "real" IP address - 80.96.148.8/32. The "upstream" router I'm talking to is 80.96.148.1. Somehow packets via the PtP interface automatically goes that direction. Anyway - the (first) problem is that bird doesn't seem to listen to port 179 (on any interfaces). In our DNS server we just do a regular listen on 80.96.148.8 (and some of the IPs on the dummy0 interface) and it's working as normal. I tried with and without "bgp listen address 80.96.148.8 port 179" and no matter what it seems like bird just doesn't on this box. I'm using 1.2.4; the box is a recent CentOS (again, under OpenVZ). Any ideas? - ask -- Ask Bjørn Hansen, http://askask.com/

2 1

default gateway
by Jaroslav Jirásek 12 Aug '10

12 Aug '10

Hi , I want to create ring network with OSPF. It works for me, but I cannot propagate default route to ring. On router with internet access I have this configuration: router id x.x.x.x protocol kernel { persist; scan time 20; import all; export all; } protocol device { scan time 10; } protocol static { route 0.0.0.0/0 via x.x.x.y; } protocol ospf MyOSPF { tick 2; rfc1583compat yes; area 10.32.0.0 { stub no; networks { a.a.a.a/30; b.b.b.b/30; }; interface "vlan10" { hello 10; retransmit 5; cost 10; transmit delay 1; dead 40; wait 40; type broadcast; authentication none; }; interface "vlan11" { hello 10; retransmit 5; cost 100; transmit delay 1; dead 40; wait 40; type broadcast; authentication none; }; }; } Other routers in ring are MikroTik routers. Thank you for any hint. Jirasek

3 2

BGP + routing w/ multiple providers/uplinks
by Gregg Berkholtz 09 Aug '10

09 Aug '10

Despite a few days of troubleshooting, I'm at a loss for answers, and Google is currently mocking further research attempts. Any help is greatly appreciated. My goals & confusion: - The intent is to announce two /24 netblocks across both of two separate uplinks, w/ my ASN via BGP. - It's not clear if I need to maintain multiple routing tables, with a single "internal" autonomous system, and if so, how to facilitate that with bird's pipe function. - I believe I should be focusing on ensuring replies return via the originating interface - however at this point, another set of eyes would really help in hashing this out. From the perspective of my router: - BGP sessions are active w/ uplinks, and providers are not filtering announcements. - ingress TCP connections to router's "source address" IP are established with no problem, if my BGP/routing tables instruct replies via the same interface packets happen to originate via. - ingress TCP connections cannot be established, if my BGP/routing tables instruct replies via a different interface than what packets happen to originate via. - tcpdump shows packets ingress from 1st provider, and egress towards 2nd provider, only if routing table instructs to 2nd provider...although replies never arrive at destination. - outbound TCP connections originating from the router itself, can be established with no issue, to either provider. Source IP matches local address of outbound route. - All addresses are public; this router is not doing any NAT. Below IPs were changed to protect the innocent, guilty, and everyone in-between - all replacements were via sed -i s///g... Since I've had no luck finding a template detailing how to accomplish this, once this hashes out, I'll followup with a relevant & documented configuration template. My bird.conf: log syslog { debug, trace, info, remote, warning, error, auth, fatal, bug }; # Avoid martians from planet RFC1918 function avoid_martians() prefix set martians; { martians = [ 169.254.0.0/16+, 172.16.0.0/12+, 192.168.0.0/16+, 10.0.0.0/8+, 224.0.0.0/4+, 240.0.0.0/4+, 0.0.0.0/32-, 0.0.0.0/0{31,32}, 0.0.0.0/0{0,7} ]; if net ~ martians then return false; return true; } protocol direct { } protocol kernel { learn; # Learn all alien routes from the kernel persist; # Don't remove routes on bird shutdown scan time 60; # Scan kernel routing table every 60 seconds import none; # Default is import all export all; # Default is export none } protocol device { scan time 60; # Scan interfaces every 60 seconds } protocol static { route 1.0.0.0/24 via "eth0"; route 2.0.0.0/24 via "eth0"; } filter bgp_out { if net ~ [2.0.0.0/24] then accept; if net ~ [1.0.0.0/24] then accept; reject; } protocol bgp INET1 { local as 11111; source address 3.0.0.118; neighbor 3.0.0.117 as 22222; import all; path metric 1; # Prefer routes with shorter paths (like Cisco does) default bgp_med 1; # MED value we use for comparison when none is defined default bgp_local_pref 100; # The same for local preference export filter bgp_out; } protocol bgp INET2 { local as 11111; source address 4.0.0.14; neighbor 4.0.0.13 as 33333; import all; path metric 1; # Prefer routes with shorter paths (like Cisco does) default bgp_med 1; # MED value we use for comparison when none is defined default bgp_local_pref 100; # The same for local preference export filter bgp_out; } As for any possible "alien" routes, the only thing I'm doing is adding the spamhaus DROP list, via this script: ip route flush table 10; ip rule del from all table 10 priority 10; curl http://www.spamhaus.org/drop/drop.lasso | sed 's/;/#/g' | sed 's/^[0-9]/ip\ route\ add\ blackhole\ &/g' | sed 's/ \#/ table 10 \#/' |sed '/^ip/!D' > spamhausDrop; ip rule add from all table 10 priority 10; sh spamhausDrop In summary, it sources http://www.spamhaus.org/drop/drop.lasso to generate a bunch of lines akin to: ip route add blackhole 95.64.98.0/23 table 10 # SBL90817 - Gregg Berkholtz Dependable IT consulting, hosting & support since 1995 www.tocici.com | 503-488-5461 | AS14613

2 2

Can't compile bird on sparc arch
by Olivier Cochard-Labbé 04 Aug '10

04 Aug '10

Hi, I'm trying to compile bird 1.2.3 under FreeBSD-sparc64. But the compilation failed with this message: cc -nostdlib -Wl,-r -o all.o cf-parse.tab.o cf-lex.o conf.o /usr/bin/ld: --relax and -r may not be used together gmake[2]: *** [all.o] Error 1 I've tried to search for a solution, and found an explanation and solution of the problem here: http://hackage.haskell.org/trac/ghc/ticket/3791 But I didn't know how to patch the bird sources files for fixing this problem. Regards, Olivier PS: full build log available here: http://pointyhat.freebsd.org/errorlogs/sparc64-errorlogs/e.8.20100704113633…

4 5

New release 1.2.4
by Ondrej Filip 03 Aug '10

03 Aug '10

Hi! New version is ready. Version 1.2.4 (2010-08-03) o Added 'show memory' command. o Important bugfix in IPv6 BGP. o Several minor bugfixes. Fedora and Debian packages are ready. I believe all other distro packages will follow soon. We are working on version 1.3.0, main change would be recursive next hop for iBGP. Ondrej

1 0