June 2010 - Bird-users - Mailing Lists

asfilter/bgpmask
by Arjan Filius 12 Jun '10

12 Jun '10

Hello bird-users, i'm trying to make a bgp_out filter statement which allows only it's own AS or multiple time's it's own AS. tried: define AS = YYY; if ( (bgp_path ~ [= AS =] || bgp_path ~ [= AS AS =] ) && net = X.X.X.X/X ) then accept; which works fine, but if not very elegant/flexible Is there a better/more elegant way to do so? Regards, Arjan Filius -- Arjan Filius mailto:iafilius@xs4all.nl

2 1

Configuring IP addresses on devices
by Tore Anderson 12 Jun '10

12 Jun '10

Hi list, I'm new to BIRD - considering making the switch from Quagga as I'm tired of bugs (in OSPFv3 especially). The first part of my existing Quagga configuration I attempted to convert, but could not figure out how to, was the following: interface eth0 ip address 192.168.0.1/24 ipv6 address 2001:db8::1/64 Am I right in understanding that there's no way of doing this in BIRD? I could keep the zebra daemon around for this purpose but it would be kinda lame... I'm also curious to hear if there's any plans for VRRP support? Best regards, -- Tore Anderson Redpill Linpro AS - http://www.redpill-linpro.com/

2 1

input filter to set outgoing prefferences
by Arjan Filius 10 Jun '10

10 Jun '10

Hello bird-users, (email send in html to prevent text wrapping, .txt version attached) running bird 1.2.3 ppa x64 ubuntu 10.04LTS. 2 bird routers connected with 2 iBGP sessions over different links. I'm trying to route specific outgoing traffic over a specific Transit. It starts with that I'm not really sure how to do it, so tested a bit. Setting Local performance on protocol definition works just fine for a complete Transit, however when i want to do it more fine grained, i run into trouble. First started with prepending routes in bgp input filter with my own AS, which works OK on a single bird router, however, with iBGP all prepended local AS-es are stripped, and that concept doesn't work. Then, started to set the preference in the bgp input filter, which also works fine on a single bird router, but that preference seems not to propagate to the second iBGP router. on router1 i set in the bgp input filter the preference to 300 for a specific network(91.189.88.0/21), and on router 2 the preference to 400. What i would want, is that both iBGP routers agree on route for a specific manual overruled prefix. below you can see each router takes its direct route to a Transit, relevant filters and show route all below. Any idea on how you want to do that? detail info: router1: bird> show route 91.189.88.0/21 91.189.88.0/21 via <Transit1_BGPRTR_1> on eth5 [Transit1_BGPRTR_1 2010-06-09 10:30:31] * (300) [AS41231i] via <iBGP_router2_gw1> on eth0 [iBGP_router1_1 2010-06-09 11:22:55] (100) [AS41231i] via <iBGP_router2_gw2> on eth1 [iBGP_router1_2 2010-06-09 11:22:55] (100) [AS41231i] via <Transit1_BGPRTR_2> on eth5 [Transit1_BGPRTR_2 2010-06-09 10:30:34] (300) [AS41231i] router2: bird> show route 91.189.88.0/21 91.189.88.0/21 via <Transit2_BGPRTR_1> on eth5 [Transit2_BGPRTR_1 2010-06-09 11:22:06] * (400) [AS41231i] via <iBGP_router1_gw1> on eth1 [iBGP_router1_1 2010-06-09 10:29:43] (100) [AS41231i] via <Transit3_BGPRTR_1> on opn-ext.382 [Transit3_BGPRTR_1 2010-06-09 11:19:12] (100) [AS41231i] via <iBGP_router1_1> on opn-ext.382 [Transit3_BGPRTR_2 2010-06-09 11:14:03] (100) [AS41231i] via <iBGP_router1_gw2> on eth0 [iBGP_router1_2 2010-06-09 10:29:43] (100) [AS41231i] bgp input filter, applied for eBGP router1: function bgp_in(int peeras) { <snip> # Traffic engineering: # launchpad.net if (net = 91.189.88.0/21 && bgp_path.first = Transit1AS ) then preference=300; # end Traffic engineering return true; } router2: function bgp_in(int peeras) { <snip> # Traffic engineering: # ppa.launchpad.net if (net = 91.189.88.0/21 && bgp_path.first = Transit2AS ) then preference=400; # end Traffic engineering return true; } router1: bird> show route 91.189.88.0/21 all 91.189.88.0/21 via <Transit1_BGPRTR_1> on eth5 [Transit1_BGPRTR_1 2010-06-09 10:30:30] * (300) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 24785 286 6908 41231 BGP.next_hop: <Transit1_BGPRTR_1> BGP.med: 100 BGP.local_pref: 100 via <iBGP_router2_gw1> on eth0 [iBGP_router1_1 2010-06-09 11:22:54] (100) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 38930 16265 286 6908 41231 BGP.next_hop: <Transit2_BGPRTR_1> BGP.local_pref: 100 BGP.community: (286,19) (286,29) (286,49) (286,286) (286,3044) (16265,65400) via <iBGP_router2_gw2> on eth1 [iBGP_router1_2 2010-06-09 11:22:54] (100) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 38930 16265 286 6908 41231 BGP.next_hop: <Transit2_BGPRTR_1> BGP.local_pref: 100 BGP.community: (286,19) (286,29) (286,49) (286,286) (286,3044) (16265,65400) via <Transit1_BGPRTR_2> on eth5 [Transit1_BGPRTR_2 2010-06-09 10:30:33] (300) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 24785 286 6908 41231 BGP.next_hop: <Transit1_BGPRTR_2> BGP.med: 100 BGP.local_pref: 100 bird> router2: bird> show route 91.189.88.0/21 all 91.189.88.0/21 via <Transit2_BGPRTR_1> on eth5 [Transit2_BGPRTR_1 2010-06-09 11:22:07] * (400) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 38930 16265 286 6908 41231 BGP.next_hop: <Transit2_BGPRTR_1> BGP.local_pref: 100 BGP.community: (286,19) (286,29) (286,49) (286,286) (286,3044) (16265,65400) via <iBGP_router1_gw1> on eth1 [iBGP_router1_1 2010-06-09 10:29:44] (100) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 24785 286 6908 41231 BGP.next_hop: <Transit1_BGPRTR_1> BGP.med: 100 BGP.local_pref: 100 via <Transit3_BGPRTR_1> on opn-ext.382 [Transit3_BGPRTR_1 2010-06-09 11:19:13] (100) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 20562 24785 286 6908 41231 BGP.next_hop: <Transit3_BGPRTR_1> BGP.med: 0 BGP.local_pref: 100 via <iBGP_router1_1> on opn-ext.382 [Transit3_BGPRTR_2 2010-06-09 11:14:04] (100) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 20562 24785 286 6908 41231 BGP.next_hop: <iBGP_router1_1> BGP.med: 0 BGP.local_pref: 100 via <iBGP_router1_gw2> on eth0 [iBGP_router1_2 2010-06-09 10:29:44] (100) [AS41231i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 24785 286 6908 41231 BGP.next_hop: <Transit1_BGPRTR_1> BGP.med: 100 BGP.local_pref: 100 bird> Regards, Arjan Filius

2 2

routes with iBGP
by Arjan Filius 09 Jun '10

09 Jun '10

Hello bird-users, Running bird 1.2.3-ppa ubuntu 10.04,x64, on 2 hosts, internally connected with 2 lines, and running 2 iBGP sessions, both routers have their own GlobalTransit provider the iBGP protocols have no filters on both routers What is noticed (for example with network 40.0.40.0/21): router1: bird> show route 40.0.40.0/21 40.0.40.0/21 via <Transit1_BGPRTR1> on eth5 [Transit1_BGPRTR1_IPv4 2010-06-04 11:26:29] * (100) [AS4249i] via <Transit1_BGPRTR2> on eth5 [Transit1_BGPRTR2_IPv4 2010-06-04 11:26:39] (100) [AS4249i] bird> router2: bird> show route 40.0.40.0/21 40.0.40.0/21 via <gw1_router1> on eth1 [iBGP_IPv4_1 2010-06-08 13:39:27] * (100) [AS4249i] via <gw2_router1> on eth0 [iBGP_IPv4_2 2010-06-08 13:28:18] (100) [AS4249i] via <Transit2_BGPRTR1> on eth5 [Transit1_BGPRTR1_IPv4 2010-06-04 11:26:30] (100) [AS4249i] bird> so router1 isn't showing it's route option to a transit on router2 (via 2 iBGP sessions) I was expecting to show all routes, but clearly it is not. I'm not sure if my expection is wrong, or not. Is this correct, end if yes why? Regards, Arjan Filius -- Arjan Filius mailto:iafilius@xs4all.nl

2 3

Re: Compiling for IPv6
by Ondrej Filip 04 Jun '10

04 Jun '10

On 28.5.2010 11:51, Simone Morandini wrote: > Hi all, > > I'd like to set up an IPv6 instance of BIRD for our members, but I'm > having some troubles with "make". > When I issue "./configure --enable-ipv6" everything goes fine, but the > subsequent "make" at a certain poing gives these errors: > > [...] > gcc -I../ -I../.. -g -O2 -Wall -Wstrict-prototypes -Wno-parentheses > -Wno-pointer-sign -o bitops.o -c bitops.c > gcc -I../ -I../.. -g -O2 -Wall -Wstrict-prototypes -Wno-parentheses > -Wno-pointer-sign -o checksum.o -c checksum.c > gcc -I../ -I../.. -g -O2 -Wall -Wstrict-prototypes -Wno-parentheses > -Wno-pointer-sign -o event.o -c event.c > event.c: In function ‘ev_run_list’: > ../lib/lists.c:132: warning: dereferencing pointer ‘({anonymous})’ does > break strict-aliasing rules > ../lib/lists.c:115: note: initialized from here > gcc -I../ -I../.. -g -O2 -Wall -Wstrict-prototypes -Wno-parentheses > -Wno-pointer-sign -o io.o -c io.c > io.c: In function ‘sysio_process_rx_cmsgs’: > io.c:704: error: dereferencing pointer to incomplete type > io.c:705: error: dereferencing pointer to incomplete type > io.c: In function ‘sk_read’: > io.c:1380: error: invalid application of ‘sizeof’ to incomplete type > ‘struct in6_pktinfo’ > make[2]: *** [io.o] Error 1 > make[2]: Leaving directory `/usr/local/src/bird-1.2.2/obj/lib' > make[1]: *** [subdir] Error 2 > make[1]: Leaving directory `/usr/local/src/bird-1.2.2/obj' > make: *** [all] Error 2 > [root@rs1 bird-1.2.2]# > > > On the same machine BIRD compiles successfully without the v6 option. Am > I missing something basic here? > The box runs on a Fedora 12. Hi Simone, this is a known bug. It is fixed differently in current git repository, but you can for example use my hot fix from Fedora RPMs. (Attached). Ondrej > > Thanks a lot, > Simone. >

2 1

linux# netstat -rn or bird> route show strangeness
by Arjan Filius 04 Jun '10

04 Jun '10

Hello, today i noticed huge difference between our 2 identical bgp ubuntu 10.04 LTS with burf 1.2.3 (ppa) x64 routers. (hardware DL380G6 12G ram X5560 @ 2.80GHz) systems are still idle in having no ip traffic yet, except for the bgp sessions and "random" other traffic. root@snxbgpprd01:~# time netstat -rn |wc -l 318259 real 0m1.296s user 0m0.950s sys 0m0.390s root@snxbgpprd02:~# time netstat -rn |wc -l 318263 real 0m9.079s user 0m6.720s sys 0m2.500s with linux ip route: root@snxbgpprd01:~# time ip route show |wc -l 318267 real 0m1.152s user 0m0.490s sys 0m1.040s root@snxbgpprd02:~# time netstat -rn |wc -l 318263 real 0m9.079s user 0m6.720s sys 0m2.500s and with bird: root@snxbgpprd01:~# time echo "show route" | birdc | wc -l 1229972 real 0m3.315s user 0m0.510s sys 0m0.230s root@snxbgpprd02:~# time echo "show route" | birdc | wc -l 896174 real 0m20.018s user 0m2.050s sys 0m0.920s so asking the route table takes at least 7 times more on the second rtr, while being exact the same hardware, OS and software. only other bird config due to IX connections. with "debug protocols { states, routes , filters};" i didn't notice anything special, and both routers seem more or less equal. with these times, i noticed that it seems that the cpu usage of bird rizes to 100% as i do a netstat -rn on the linux prompt. It looks like a route table access contention At one time, this morning, the netstat command exceded the 3 minutes, and i finally killed it. Has anyone same experinces, or knows perhaps what is going on? i want to start with oprofile to have a more closer look what is going on, and where that cpu time is consumed. Regards, -- Arjan Filius mailto:iafilius@xs4all.nl

1 1

Small bug in Redhat init script
by Neil Wilson 02 Jun '10

02 Jun '10

Hi, There appears to be a small bug in the IPv6 section of the init script. [ -f /etc/bird-6.conf ] || BIRD6="no" should be [ -f /etc/bird6.conf ] || BIRD6="no" Similarly on the default control file within birdc6 - which won't attach. Unable to connect to server control socket (/var/run/bird-6.ctl): No such file or directory

2 1

New release 1.2.3
by Ondrej Filip 02 Jun '10

02 Jun '10

Hi! New version is ready. Version 1.2.3 (2010-06-01) o Pattern matching for community lists. o Many fixes in OSPF protocol (esp. in multi-areas and vlinks). o Several minor bugfixes. Fedora and Debian packages are ready. I believe all distro packages will follow soon. Ondrej

2 2