January 2013 - Bird-users

Re: BUG: /lib/lists.c.. Don`t use -O2 . Use -O1
by Uwe Kleine-König 22 Jan '13

22 Jan '13

On Tue, Jan 22, 2013 at 04:20:06PM +0400, Maksim Tamarsky wrote: > I don`t exprt in ARM-ASM but my colleague say and i se this, that some > ASM code from rem_node and add_head was interlaced and will now be run > in parallel mode and this can lead to error. It's normal that the compiler does reorder the machine instructions for two consecutive inline functions. The overall result still is the same, at least if you stick to defined C constructs and your compiler doesn't have a bug. I don't understand what you mean saying "parallel mode", AFAIK bird is single-threaded so there is nothing running in parallel. If your colleague still thinks there is a bug please ask him to describe it here on the mailing list. Best regards Uwe

1 0

BUG: /lib/lists.c
by Тамарский Максим 21 Jan '13

21 Jan '13

LIST_INLINE void rem_node(node *n) { node *z = n->prev; node *x = n->next; z->next = x; x->prev = z; } Using of this function may broke list.tail and list.head. Later call to function like add_head or add_tail may couse problem. -- Тамарский Максим Александрович Инженер-программист ООО Эн-Эс-Джи г.Москва, Кирпичная 39/41 оф.1302 Почтовый адрес: 105187 Россия, Москва ул.Вольная, д.35, оф.NSG Тел./факс: +7 495 727-19-59 Web-сайт: http://www.nsg.ru

6 6

Seems I can't get bird to open a bgp listen socket
by Hans van Kranenburg 21 Jan '13

21 Jan '13

Hi Bird users, I'm setting up a bird test router, using bird 1.3.9 on debian squeeze (backport of the packages in unstable). It seems I cannot get bird to listen on tcp port 179. tcpdump shows outgoing traffic from 10.140.59.20 to 10.140.59.16, but there's just nothing listening on port 179 when I start bird. $ netstat -an | grep 179 $ This really feels like I'm missing something small, something stupid. Here's my test config: router id 10.140.59.20; debug protocols all; protocol kernel { persist; scan time 60; import none; # filter!!1~1 export all; } protocol device { scan time 60; } protocol static { import all; route 10.0.0.0/8 blackhole; route 172.16.0.0/12 blackhole; route 192.168.0.0/16 blackhole; debug all; } protocol ospf ospf1918 { area 0.0.0.0 { interface "eth2" { authentication cryptographic; password "xxxx" { id 25; }; }; interface "eth1" { stub; }; interface "lo" { stub; }; }; } protocol bgp bgp1918_knud { import all; local as 64521; source address 10.140.59.20; keepalive time 20; hold time 60; neighbor 10.140.59.16 as 64521; } protocol bgp bgp1918_oele { import all; local as 64521; source address 10.140.59.20; keepalive time 20; hold time 60; neighbor 10.140.59.18 as 64521; } I tried to explicitely add "listen bgp address 0.0.0.0 port 179;" to the config, but still no listening 179. Logs do not show any errors or warnings why the listening socket is not opened... Jan 20 14:29:12 cabbage bird: bgp1918_knud: Initializing Jan 20 14:29:12 cabbage bird: bgp1918_oele: Initializing Jan 20 14:29:12 cabbage bird: bgp1918_knud: Starting Jan 20 14:29:12 cabbage bird: bgp1918_oele: Starting Jan 20 14:29:12 cabbage bird: bgp1918_oele: Waiting for 10.140.59.18 to become my neighbor Jan 20 14:29:12 cabbage bird: bgp1918_knud: Waiting for 10.140.59.16 to become my neighbor Here's some output of show protocols. bird> show protocols all bgp1918_oele name proto table state since info bgp1918_oele BGP master start 14:19 Idle Preference: 100 Input filter: ACCEPT Output filter: REJECT Routes: 0 imported, 0 exported, 0 preferred Route change stats: received rejected filtered ignored accepted Import updates: 0 0 0 0 0 Import withdraws: 0 0 --- 0 0 Export updates: 0 0 0 --- 0 Export withdraws: 0 --- --- --- 0 BGP state: Idle Neighbor address: 10.140.59.18 Neighbor AS: 64521 -- Hans van Kranenburg - System / Network Engineer T +31 (0)10 2760434 | hans.van.kranenburg(a)mendix.com | www.mendix.com

2 4

BIRD - send notification event when session is UP
by Javor Kliachev 21 Jan '13

21 Jan '13

2 2

bird6 1.3.7 hangs on start on a Raspberry Pi
by Eric Buddington 18 Jan '13

18 Jan '13

The config below causes bird6 to hang, without backgrounding, on my Raspberry Pi (model B). The same config on a vortex96sx system (also bird6-1.3.7 under Debian) causes bird6 to apparently start normally - at least it forks and returns. strace on the Pi shows a couple of sendmsg/recvmsg transactions (no errors), then silence. bird6's CPU usage pegs. How can I debug this further? -Eric ------------------------ bird6.conf ------------------- router id 0.0.0.2; protocol kernel { } protocol device { scan time 10; # Scan interfaces every 10 seconds } protocol static { } protocol ospf MyOSPF { area 0 { interface -"lo" { type broadcast; }; }; }

4 5

RIP over unnumbered links
by Gustavo Ponza 18 Jan '13

18 Jan '13

Hi developers, confirm again the very good work on BIRD 1.3.9 and we, amateur radio fellows, are very enthusiast for all implemented features :) Since I am receiving on my linux-box 'ir0rm-7.ampr.org' the RIP announces of the living ampr.org packet radio gateways by the main RIPv2 router operating at ucsd.edu (California), then activated the RIP on the BIRD to 'catch' and manage through my radio amateur 'pseudo-router' that gateway routing information but obtain the following log info: Jan 17 17:41:52 ir0rm-7 bird: rip1: rip is not defined over unnumbered links Is it actually possible to *define* the BIRD RIP to manage that stuff, and how? Or, if not, due to the RIP design: is it possible to have that feature implemented on the next BIRD release? Many thanks again for all gus

1 0

New release 1.3.9
by Ondrej Filip 18 Jan '13

18 Jan '13

Hi Bird users, we have a new version of your popular routing daemon. The major improvement is related to filters. Many thanks to Santiago. Ondrej Version 1.3.9 (2013-01-11) o BIRD can be configured to keep and show filtered routes. o Separate receive and import limits. o Several new reconfiguration cmd options (undo, timeout, check). o Configurable automatic router ID selection. o Dragonfly BSD support. o Fixed OSPFv3 vlinks. o Several minor bugfixes

5 5

Configuration Problem : announce default route via ospf
by Raphael Mazelier 17 Jan '13

17 Jan '13

Hello list, I'm a new user of bird. I'm using quagga or openospdf in production for time (also cisco and junos). Bird have an excellent reputation and a small footprint. I m trying to replace my old quagga on embedded box (running openwrt) with bird. The configuration look very flexible but I'm facing with a problem. What I try to make is announcing a default route in ospf, and also set some specific metric. I'm lost with that. The quagga/ospfd relevant part of configuration I try to make : - snip - interface lo2 ip ospf cost 100 router ospf ospf router-id 0.0.0.9 redistribute connected route-map connected_to_ospf passive-interface lo0 passive-interface lo1 passive-interface lo2 network 192.168.0.0/24 area 0.0.0.5 network 192.168.10.9/32 area 0.0.0.5 network 192.168.10.53/32 area 0.0.0.5 default-information originate metric 100 ! access-list access permit 127.0.0.1/32 access-list access deny any ! route-map connected_to_ospf deny 10 match interface mpe0 ! route-map connected_to_ospf deny 15 match interface vlan44 ! route-map connected_to_ospf permit 20 match interface lo2 set metric 100 set metric-type type-1 - /snip - I gave up with a config like this : - snip - router id 0.0.0.9; protocol device { scan time 10; } filter export_OSPF { if ( net = 0.0.0.0/0 ) then { print "default net accepted:", net; ospf_metric1 = 100; accept; } accept } protocol kernel { debug all; learn; scan time 2; import all; export all; } protocol ospf { debug all; import all; export filter export_OSPF; area 0.0.0.5 { stubnet 192.168.10.53/32 { cost 100; }; interface "lo1" { stub; }; interface "lo2" { stub; }; interface "vr0" { cost 5; type broadcast; }; interface "vr1" { stub; cost 200; }; }; } - /snip - It's not working as expected. In fact I'm trying to redistribute defautl kernel route into ospf. but it seems to be ignored, and the default was not seeen in the ospf database. What I am missing ? Is there a better way to inject a default route like in quagga (default originate). Regards, -- Raphael Mazelier

3 4

Re: Route aggregation in BIRD - how?
by Ondrej Zajicek 16 Jan '13

16 Jan '13

On Wed, Jan 09, 2013 at 04:12:48AM +0400, Alexander V. Chernikov wrote: > Version 7 :) > Changes > * Fix bug with using 0/0 and :: in aggregation protocol > * Fix mandatory lists not working before second rehash > * Improve docs and configuration example a bit > * Some code cleanups > > Thanks to Robers Hass for his bugreport and testing. Hello I would like to merge the aggregator, but i have three related conceptual problems with your patch: (1) About third of the code is related to BGP attribute processing according to RFC 4271 9.2.2.2 (esp. AS_PATH merging and AS_SET construction) for 'save attributes' option, but this behavior is IMHO mostly useless, as such behavior is deprecated, RFC 6472 recommends that AS_SETs should not be generated and there is a RFC draft [*] that even prohibits them. As this feature adds significant and unnecessary complexity, i would like to remove it. [*] http://tools.ietf.org/html/draft-kumari-deprecate-as-set-confed-set-01 (2) If i understand it correctly, your aggregator keeps a copy of all received and aggregated routes in an internal fib, which is useful for implementing (1), but unnecessarily if you just want to originate an aggregate route if there are some matching routes (or required number of mandatory routes). In that case you just need some counters for summary routes and flags for mandatory routes. Current implementation is especially memory wasteful in the common case 'aggregate full BGP feed to get a default route'. (3) Generated aggregate routes have attrs->proto->proto == proto_agg, but attrs->source == RTS_BGP. What is a reason for that? It does not make much sense to me, because it does not have true BGP route behavior (for example rte_better would not compare it with others because of attrs->proto->proto) so i see no reason why not to define RTS_AGGREGATOR and use it. Even if we would like to generate BGP-style aggregated routes with BGP_AGGREGATOR and BGP_ATOMIC_AGGR attributes. There are some minor issues (e.g. there is probably a bug in bgp_update_sumroute() in BGP_ATOMIC_AGGR generation - even if not received from aggregated routes, this attribute should be generated if BGP routes with non-empty AS_PATH are aggregated and 'save attributes' is disabled), but these probably aren't problematic. If you do not have strong objections, i would remove (1), do some readjustment for (2), fix some minor issues and merge it. Question is whether there is any need to have optional keeping a copy of child routes in (2) for some sophisticated aggregation modes like (1), but without (1), i currently don't see the need. -- 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 2

OSPF RFC compliance issue?
by Aragon Gouveia 11 Jan '13

11 Jan '13

Hi birdsters, I'm having difficulty getting a bird OSPF instance to peer correctly with a MikroTik RouterOS device over an L2TP PTP link. The symptom is that routes originated from my bird instance show up in RouterOS's route database with no gateway, and are not added to the device's routing table as a result. MikroTik have debugged this and claim that bird's OSPF LS updates are violating RFC. After reading a packet capture and the relevant parts of RFC 2328, they seem to be correct. I've provided a PCAP capture at: http://decoder.geek.sh/misc/bird-mt-ospf.pcap The interesting packet is #14 in the capture. Bird is sending a Router LSA with bizarre link information that does not correspond at all with the PTP link information. In reality the PTP link on the bird device has address 172.18.102.37 with a peer address of 172.18.102.36, and these addresses are unique across all interfaces. I'm using bird 1.3.8 on Debian Wheezy (Linux kernel 3.2.32). Any ideas? Thanks, Aragon

2 2