July 2011 - Bird-users - Mailing Lists

All protocols restarting on configure soft
by Stephen Holmes 28 Jul '11

28 Jul '11

BIRD Masters, BIRD 1.2.5 I recently updated my route server configuration to include a new set of BGP export filters when one peer wants to exclude routes to another peer. The filter is only applied to those selecting to exclude one another. When I pushed the new config with only two BGP sessions having the added filers, the rest untouched, and two new pipes were added, BIRD restarted all PIPE/BGP sessions after a 'configure soft'. Is this the expected behavior? Without export filter: protocol bgp RXXXXX { local as myas; neighbor xx.xx.xx.xx as XXXXX; import filter bgp_in_ASXXXXX; export all; route limit 500; table TXXXXX; rs client; start delay time 30; } With export filter: protocol bgp RYYYY { debug { filters }; local as myas; neighbor yy.yy.yy.yy as YYYY; import filter bgp_in_ASYYYY; export filter bgp_out_ASYYYY; route limit 1000; table TYYYY; rs client; start delay time 30; } Much Appreciated, Stephen

1 2

init script
by Ryan Whelan 27 Jul '11

27 Jul '11

Looking through the source, I don't see any init scripts for controlling the starting and stopping of Bird. Before I write something, I wanted to check if they already existed and I'm just missing them. Thanks

4 4

Advertising interface addresses via OSPF
by Ryan Whelan 27 Jul '11

27 Jul '11

I have 3 Linux machines connected to one another by point-to-point links. With the end point of each tunnel using a tun device, there are a total of 6 tun devices used on across the 3 machines. 2 per machine. Using OSPF, Bird is not advertising the addresses on the tun devices to the other nodes. This causes an issue because each device is unaware of the addresses on the tun devices on tunnels it is not connected to. In other words, node 'C' is unaware of the addresses on the tunnel between nodes 'A' and 'B'. Thats an issue because if the link between 'A' and 'B' fails, 'A' will get the routes 'B' knows about thru 'C', but 'B' does not know the tun address of 'A' thats connected to 'C' so all the traffic 'C' routes from 'A' to 'B' comes from an address 'B' can't route back too. (Sorry its so hard to follow) I've been able to alleviate it by adding a `stubnet` for each tun device on each machine, but I'm afraid that will become increasingly difficult to maintain as the number of nodes increases. Is there another way? I've been unsuccessful at getting bird to import (and advertise) the routes associated with the tun devices. (using the 'device' protocol I can get bird to import them, but ospf doesn't advertise them to the other nodes)

2 6

strange behaviour with direct protocol if two interface has the same ip address
by csszep 25 Jul '11

25 Jul '11

Hello! Openswan with klips stack use ipsec* interface which is configured to the same ip address as the physical interface. If i use the direct protocol and starting ipsec daemon the physical interface's connected route disappears. If i stopping ipsec daemon any connected route is missing from direct protocol. Simple test: bird.conf: debug protocols all; protocol direct { } protocol kernel { learn; persist; scan time 20; export all; } protocol device { scan time 10; } protocol static { } Ipsec stopped and after that bird started: BIRD 1.3.2 ready. bird> show interfaces lo up (index=1) MultiAccess AdminUp LinkUp Loopback Ignored MTU=16436 127.0.0.1/8 (Primary, scope host) eth0 up (index=2) MultiAccess Broadcast Multicast AdminUp LinkUp MTU=1500 10.0.2.15/24 (Primary, scope site) eth1 up (index=3) MultiAccess Broadcast Multicast AdminUp LinkUp MTU=1500 192.168.56.3/24 (Primary, scope site) ipsec0 DOWN (index=7) MultiAccess AdminDown LinkDown MTU=16260 ipsec1 DOWN (index=8) MultiAccess AdminDown LinkDown MTU=0 mast0 DOWN (index=9) MultiAccess AdminDown LinkDown MTU=0 bird> sho route 0.0.0.0/0 via 10.0.2.2 on eth0 [kernel1 16:16] * (10) 10.0.2.0/24 dev eth0 [direct1 16:16] * (240) 192.168.56.0/24 dev eth1 [direct1 16:16] * (240) Ipsec started: bird> show interfaces lo up (index=1) MultiAccess AdminUp LinkUp Loopback Ignored MTU=16436 127.0.0.1/8 (Primary, scope host) eth0 up (index=2) MultiAccess Broadcast Multicast AdminUp LinkUp MTU=1500 10.0.2.15/24 (Primary, scope site) eth1 up (index=3) MultiAccess Broadcast Multicast AdminUp LinkUp MTU=1500 192.168.56.3/24 (Primary, scope site) ipsec0 up (index=7) MultiAccess AdminUp LinkUp MTU=16260 10.0.2.15/24 (Primary, scope site) ipsec1 DOWN (index=8) MultiAccess AdminDown LinkDown MTU=0 mast0 DOWN (index=9) MultiAccess AdminDown LinkDown MTU=0 bird> sho route 0.0.0.0/0 via 10.0.2.2 on eth0 [kernel1 16:16] * (10) 10.0.2.0/24 dev ipsec0 [direct1 16:19] * (240) <-------- connected route changed to ipsec0 192.168.56.0/24 dev eth1 [direct1 16:16] * (240) Ipsec stopped again: bird> show interfaces lo up (index=1) MultiAccess AdminUp LinkUp Loopback Ignored MTU=16436 127.0.0.1/8 (Primary, scope host) eth0 up (index=2) MultiAccess Broadcast Multicast AdminUp LinkUp MTU=1500 10.0.2.15/24 (Primary, scope site) eth1 up (index=3) MultiAccess Broadcast Multicast AdminUp LinkUp MTU=1500 192.168.56.3/24 (Primary, scope site) ipsec0 DOWN (index=7) MultiAccess AdminDown LinkDown MTU=16260 ipsec1 DOWN (index=8) MultiAccess AdminDown LinkDown MTU=0 mast0 DOWN (index=9) bird> show route 0.0.0.0/0 via 10.0.2.2 on eth0 [kernel1 16:16] * (10) 192.168.56.0/24 dev eth1 [direct1 16:16] * (240) Connected route from eth0 is missing. Log: Jul 23 16:31:19 debianvm1 ipsec_setup: Starting Openswan IPsec 2.6.28... Jul 23 16:31:19 debianvm1 ipsec_setup: Using KLIPS/legacy stack Jul 23 16:31:20 debianvm1 ipsec_setup: KLIPS debug `none' Jul 23 16:31:20 debianvm1 ipsec_setup: KLIPS ipsec0 on eth0 10.0.2.15/255.255.255.0 broadcast 10.0.2.255 Jul 23 16:31:20 debianvm1 bird: static1 < interface ipsec0 goes up Jul 23 16:31:20 debianvm1 bird: direct1 < primary address 10.0.0.0/8 on interface ipsec0 added Jul 23 16:31:20 debianvm1 bird: direct1 > added [best] 10.0.0.0/8 dev ipsec0 Jul 23 16:31:20 debianvm1 bird: kernel1 < rejected by protocol 10.0.0.0/8 dev ipsec0 Jul 23 16:31:20 debianvm1 bird: direct1 < primary address 10.0.0.0/8 on interface ipsec0 removed Jul 23 16:31:20 debianvm1 bird: direct1 > removed [sole] 10.0.0.0/8 dev ipsec0 Jul 23 16:31:20 debianvm1 bird: static1 < interface ipsec0 goes down Jul 23 16:31:20 debianvm1 ipsec_setup: ...Openswan IPsec started Jul 23 16:31:20 debianvm1 ipsec__plutorun: adjusting ipsec.d to /etc/ipsec.d Jul 23 16:31:20 debianvm1 ipsec__plutorun: 003 NAT-Traversal: Trying new style NAT-T Jul 23 16:31:22 debianvm1 bird: device1: Scanning interfaces Jul 23 16:31:22 debianvm1 bird: static1 < interface ipsec0 goes up Jul 23 16:31:22 debianvm1 bird: direct1 < primary address 10.0.2.0/24 on interface ipsec0 added Jul 23 16:31:22 debianvm1 bird: direct1 > added 10.0.2.0/24 dev ipsec0 Jul 23 16:31:22 debianvm1 bird: kernel1 < rejected by protocol 10.0.2.0/24 dev ipsec0 Jul 23 16:31:32 debianvm1 bird: device1: Scanning interfaces Jul 23 16:31:32 debianvm1 bird: kernel1: Scanning routing table Jul 23 16:31:32 debianvm1 bird: kernel1: 0.0.0.0/0: [alien] seen Jul 23 16:31:32 debianvm1 bird: kernel1: Pruning table master Jul 23 16:31:32 debianvm1 bird: kernel1: Pruning inherited routes Jul 23 16:31:35 debianvm1 ipsec_setup: Stopping Openswan IPsec... Jul 23 16:31:37 debianvm1 bird: direct1 < primary address 10.0.2.0/24 on interface ipsec0 removed Jul 23 16:31:37 debianvm1 bird: direct1 > removed [sole] 10.0.2.0/24 dev ipsec0 Jul 23 16:31:37 debianvm1 bird: static1 < interface ipsec0 goes down Jul 23 16:31:37 debianvm1 ipsec_setup: ...Openswan IPsec stopped Jul 23 16:31:41 debianvm1 bird: device1: Scanning interfaces Jul 23 16:31:51 debianvm1 bird: device1: Scanning interfaces Jul 23 16:31:51 debianvm1 bird: kernel1: Scanning routing table Jul 23 16:31:51 debianvm1 bird: kernel1: 0.0.0.0/0: [alien] seen Jul 23 16:31:51 debianvm1 bird: kernel1: Pruning table master Jul 23 16:31:51 debianvm1 bird: kernel1: Pruning inherited routes thx Csszep

2 1

bird configure error
by Ryan Whelan 19 Jul '11

19 Jul '11

Im getting the error "configure: error: The client requires GNU readline library 2.1 or newer. Either install the library or use --disable-client to compile without the client." from the configure script on a CentOS 5.6 64bit machine. I have the the readline RPM from the distribution repository as well as the distributions development package: # rpm -qa | grep readline readline-devel-5.1-3.el5 readline-5.1-3.el5 searching the readline header files returns what I think the configure script is checking for: # grep rl_callback_read_char /usr/include/readline/* /usr/include/readline/readline.h:extern void rl_callback_read_char PARAMS((void)); Adding '-I/usr/include/readline' to the CFLAGS doesn't help. I feel like I'm missing something obvious but not sure where to turn. Google has not proven very fruitful. Thanks ryan

1 0

Slow static reconfigure
by Neil Wilson 14 Jul '11

14 Jul '11

Hi all, Loading a large set of prefixes into the static protocol is pretty quick, but reconfiguring is really slow. The current implementation uses linked lists which are traversed lots of times during the reconfiguration. I was wondering if these could be indexed and sorted in some way with one of the existing bird data structures - perhaps with a FIB - which would allow the use of a faster merge algorithm. Any other suggestions as to how the static reconfigure could be made faster? Rgs Neil

3 5

ibgp bird
by Gianluigi Salerno 14 Jul '11

14 Jul '11

Hi, I've a problem using IBGP, it seems that next hop attribute is lost. If I use version 1.2.5 it works fine but with 1.3.0, 1.3.1 and 1.3.2 the problem persist. Here it is the configuration on cisco side: neighbor 195.36.1.7 remote-as 5392 neighbor 195.36.1.7 next-hop-self neighbor 195.36.1.7 maximum-prefix 10 and bird: router id 195.36.1.7; protocol kernel { persist; scan time 20; export all; } protocol device { scan time 10; } protocol static { } protocol bgp milano3 { import all; local as 5392; next hop self; neighbor 195.36.1.3 as 5392; } Any hint ? Thank you. Regards. -- Gianluigi Salerno - TELNET S.r.L. Phone: +39-0382-529751 TELNET e' associata ad AIIP, MIX, RIPE Fax: +39-0382-528074 I-27100 V. Buozzi, 5 - Pavia, Italy e-mail: gsalerno(a)telnetwork.it

2 1

show protocols all output
by Marcin Kurek 12 Jul '11

12 Jul '11

Hello, I wonder if there is a simple way to force bird to display informations like ASN, neighbor address even if the session is down? Currently I use "show protocols all" but maybe there is a better way? -- Kind regards, Marcin Kurek

2 2

bird 1.3.1 protocol radv issue
by Bogdan Besliu 11 Jul '11

11 Jul '11

Hello, I'm testing bird 1.3.1 protocol radv . The configuration is: router-id 1.1.1.1; protocol device { scan time 10; }; protocol radv { interface "ppp*" prefix 2001:1111:1111::/64 { autonomous on; }; }; most of the time it works ok but from time to time it happens something like this: un 17 18:19:41 ppp-test pppd[8032]: No response to 3 echo-requests Jun 17 18:19:41 ppp-test pppd[8032]: Serial link appears to be disconnected. Jun 17 18:19:41 ppp-test pppd[8032]: Connect time 436.5 minutes. Jun 17 18:19:41 ppp-test pppd[8032]: Sent 1307925176 bytes, received 252750 bytes. Jun 17 18:19:41 ppp-test bird6: radv1: Socket error: Network is unreachable Jun 17 18:19:41 ppp-test bird6: radv1: Socket error: Network is unreachable Jun 17 18:19:41 ppp-test bird6: radv1: Socket error: Network is unreachable Jun 17 18:19:41 ppp-test bird6: radv1: Socket error: Network is unreachable ........... and the last messages repeats itself until hdd is full; or bird6 daemon stop working. it looks like the problems appear when radv tries to send RA and at the same time ppp interface goes down. any ideas ? thx, Bogdan Besliu

2 2

Notabiliy references for BIRD
by Ondrej Zajicek 11 Jul '11

11 Jul '11

Hello It seems that Wikipedia article about BIRD [*] lacks reliable references from secondary sources to establish notability of BIRD. Do you now any such references? [*] http://en.wikipedia.org/wiki/Bird_Internet_routing_daemon -- 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."

2 1