- Bird-users - Mailing Lists

Soft routes in bird
by Amadeus 26 May '12

26 May '12

Hi, is it possible to configure bird in a way that it stops advertising route A as soon as it recognizes that the same route A is now advertised by another router? So to speak, make it 'respectful' to the advertisements of others - despite the fact that it has entries of type 'static' matching A? Best wishes, Amadeus

3 2

Auto generate prefix-list
by Denis Samsonov 22 May '12

22 May '12

Dear Colleagues, sorry for offtop. Who makescript to automatically generationa bird prefix-list by AS-SET from RIPE-DB? Can you show examples?

2 1

real broadcast mode
by Tapio Haapala 16 May '12

16 May '12

Hello, Bird supports with ospf type broadcast. Looks that it actually means hat you execpt from "type mcast". On several networks multicast is broken. Yes it is broken and we cannot do nothing for that because they are metro ethernet and odsl (operator dsl) networks and they are leased from local operators. One solution is make mtmp configuration but if you have 100 router it is nightmare. So I humbly asking new feature. It is real broadcast type. Because of compatibility reasons it is unpossible use type bcast for that and rename multicast for type mcast I suggest that new type is rbcast. It is extension over rfc but it is really usefull one. On real broadcast packets send to 255.255.255.255 so they does not get meshed up with multicast specific rules. I have hear many times "but multicast is broadcast on l2 networks". Pleas dont use that argument. Truck is car, pickup is car but it does not mean that truck is pickup. Some times you have roads where you can drive with pickups but not with trucks. First time when I face this problem I was thiking that "ok this is corner case and solution is ptmp". When we start get these cases on several networks with several operators and several vendors, I start thinking that maybe there is improvement need on ospf. ps How many elese had suffer problems with metro ethernet or dsl network that devices delivery unicast traffic and bcast traffic to ff:ff:ff:ff:ff:ff but does not delivery traffic to 01:00:5e:00:00:05 -- F-Solutions Oy Tapio Haapala PL 7, 90571 Oulu GSM 040-0998371 Skype burner- IRC Burner@ircnet

3 5

Bird build on embedded system
by Bill Siadak (bsiadak) 15 May '12

15 May '12

Hello, Bird maintainers. I am having an issue when cross-compiling bird for an embedded system. The build fails due to an issue in the install where there appears to not be an option in the configure to specify a non-native strip utility to the install utility. Please provide a configure option that allows specification of a ""--strip-program=<program>" to the Makefile or remove the -s from the make install and let the strip be done later as a separate step. Thanks, Bill

1 0

Possible bug in ospf_lsupd_receive
by Benjamin Paterson 14 May '12

14 May '12

Hi guys, I think there's an endianness problem in ospf_lsupd_receive. On the lines where the program checks lsa->type or uses the LSA_SCOPE macro, it has not previously converted lsa->type to host representation. I saw this problem as I created a new LSA class with type BFF0, which has area scope. It was incorrectly displaying "<WARN> Received LSA with invalid scope" messages and refusing to add the LSAs to the database. I checked and at the line 513, if ((LSA_SCOPE(lsa) == LSA_SCOPE_RES)) was evaluating as true because lsa->type was F0BF, which is a reserved type. All the other LSA types are also inverted, but it doesn't pose a problem for the common LSA types as their first hex digit is usually 0 if they are misrepresented (0800 for link LSA, 0120 for router LSA etc). The problem can be fixed by putting ntohlsah(lsa, &lsatmp); before the checks and using &lsatmp instead of lsa for checks. Cheers

2 1

Interface Instance IDs in OSPFv3
by Benjamin Paterson 14 May '12

14 May '12

I am intrigued by the following bit of code in BIRD 1.3.7 proto/ospf/iface.c void ospf_ifa_notify(struct proto *p, unsigned flags, struct ifa *a) { [...] if (flags & IF_CHANGE_UP) { int done0 = 0; struct ospf_area *oa; WALK_LIST(oa, po->area_list) { int iid = 0; struct ospf_iface_patt *ip; while (ip = ospf_iface_patt_find(oa->ac, a->iface, iid)) { ospf_iface_new(oa, a, ip); if (ip->instance_id == 0) done0++; iid = ip->instance_id + 1; } } I might be wrong here, but if I understand correctly, the code expects the set of instance IDs configured on each interface to be of the form [| 0, n |]. In addition, I haven't seen any option to manually set Instance IDs. Is there any reason why this should be the case? I would like to manually set the instance ID to something other than 0, without necessarily having other protocol instances running on the same interface. Thanks - Benjamin

2 4

Cannot build with --enable-ipv6 on master branch
by Benjamin Paterson 11 May '12

11 May '12

At commit 95616c820248018f4999972cad315f2da60e4960 (4 May 2012) I get an error towards the end of the build process when running make after ./configure --enable-ipv6 gcc -o ../bird nest/all.o filter/all.o proto/bgp/all.o proto/ospf/all.o proto/pipe/all.o proto/radv/all.o proto/rip/all.o proto/static/all.o conf/all.o lib/birdlib.a -lrt proto/bgp/all.o: In function `bgp_do_rx_update': /home/bpaterso/git/bird-homenet.git/obj/proto/bgp/../../../proto/bgp/packets.c:1035: undefined reference to `bgp_apply_limits' collect2: ld returned 1 exit status make[1]: *** [../bird] Error 1 make[1]: Leaving directory `/home/bpaterso/git/bird-homenet.git/obj' make: *** [all] Error 2 I'm not getting this error in 1.3.7. I would love to be able to merge the most recent changes into my development branch, but right now that's keeping me from it :( Cheers

2 2

BIRD 1.3.7
by Gustavo Ponza 09 May '12

09 May '12

Hi everyone, developers and users, having played (for the *first* time) the last three months with the 'quagga' suite, incidentally, just some days ago, I discovered the BIRD program and want to try it... the router compiled and run OK on my (production) 2.6.21 linux kernel platform. So, since then, I installed it and now I'm learning and practicing for the best working configuration, focused for now, only on the 'ospf' environment in which I'm concerned. Since I experienced (at least it is for me) that the newest quagga releases doesn't work on my alternate 2.4 linux platform (kernel 2.4.23), I tried the BIRD on that and discovered it run OK... the ospf routing processes seem regular without any problems, but: 1) at launch of bird appears on console the warning: 'bird: Monotonic timer is missing' furthermore, the *syslog* reports (from 'startup' to 'shutdown') the following lines: May 6 08:25:31 ir0rm-7 bird: Monotonic timer is missing May 6 08:25:44 ir0rm-7 bird: Netlink: File exists May 6 08:26:33 ir0rm-7 last message repeated 3 times May 6 08:27:33 ir0rm-7 last message repeated 3 times May 6 08:28:33 ir0rm-7 last message repeated 3 times May 6 08:29:33 ir0rm-7 last message repeated 3 times ... ... May 6 08:49:37 ir0rm-7 bird: Netlink: No such process May 6 08:49:37 ir0rm-7 bird: Shutdown completed instead, the linux 'syslog' for 2.6 kernel reports (from 'startup' to 'shutdown') only the following lines: May 6 08:25:44 ir0rm-7 bird: Netlink: File exists May 6 08:26:33 ir0rm-7 last message repeated 3 times May 6 08:27:33 ir0rm-7 last message repeated 3 times May 6 08:28:33 ir0rm-7 last message repeated 3 times May 6 08:29:33 ir0rm-7 last message repeated 3 times May 6 08:30:33 ir0rm-7 last message repeated 3 times ... ... May 6 08:49:37 ir0rm-7 bird: Shutdown completed Now, reading the past Aug/Sep 2010 bird mailing lists, a similar problem appears to be solved. So, my question about the above warnings: are they due to my systems or to my missed/wrong setups; or instead a suddenly re-appearing of such a warnings/(failure for 2.4?) messages? No logging lines are setup on both 'bird.conf' files. TNX for your big efforts on improving the BIRD... and for your kindly reply. gus i0ojj

2 2

Blackhole routes using a filter?
by Dan Luedtke 09 May '12

09 May '12

Hi everyone, I am stuck with bird, could you please give me a hint? The setup: My router peers with Team Cymru to get fullbogons via BGP. I want to blackhole these routes using a filter. My filter looks like this: filter blackhole { gw = 2001:db8::1; accept; } However, the kernel refuses to import my blackholed routes: May 2 10:27:08 gw bird6: cymru1 > added [best] 2001:16a1::/32 via 2001:db8::1 on eth0 May 2 10:27:08 gw bird6: kernel1 < added 2001:16a1::/32 via 2001:db8::1 on eth0 May 2 10:27:08 gw bird6: cymru1 < rejected by protocol 2001:16a1::/32 via 2001:db8::1 on eth0 Any ideas how to accomplish blackholing? Other approaches maybe? Not that I like Cisco very much, but it is easier on their equipment :/ Here is the actual peering, just for the case it matters: protocol bgp cymru1 { description "Cymru IPv6 fullbogons #1"; local as 57821; neighbor 2620:0:6B0::26E5:4207 as 65332; source address 2001:67c:26f4::1; password "got lost during mail transfer somehow :)"; multihop 20; import filter blackhole; export all; } Regards, Dan -- Dan Luedtke http://www.danrl.de

4 5

No supported AFI/SAFI
by Simone Morandini 05 May '12

05 May '12

Hello list, as per the subject, I have a specific peer whose BGP session is not stable. It goes from Established to Idle with this error from the birdc prompt: "Received: No supported AFI/SAFI". The logfile says: 03-05-2012 22:09:47 <TRACE> <peername>: State changed to up 03-05-2012 22:09:50 <RMT> <peername>: Received: Malformed attribute list 03-05-2012 22:09:50 <TRACE> <peername>: State changed to flush 03-05-2012 22:09:50 <TRACE> <peername>: State changed to stop 03-05-2012 22:09:50 <TRACE> <peername>: Starting and then 03-05-2012 22:27:03 <RMT> <peername>: Received: No supported AFI/SAFI: 000101 What does this refers to? Version of BIRD is 1.3.1 on this machine. Thanks, Simone. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.

2 1