July 2010 - Bird-users - Mailing Lists

device routes switch in kernel protocol doesn't work
by Sebastian Nieszwiec [POLNET] 10 Jul '10

10 Jul '10

Hi all, I'm trying to propagate my local routes from main table to another kernel table with bird. I need it to use policy base routing. Routes are in bird table but they doesn't propagate with kernel protocol. I've found in manual "device routes switch" but it doesn't work. birdc after configure command writes parse error. I use bird-1.2.3 on debian etch. Thanks for answers. -- pozdrawiam Sebastian Nieszwiec ------------------------------------------------------------ --- FORCEnet, Komputery, Internet -------------------------- --- LAN/WAN Administrator ---------------------------------- --- email: sed -e 's/@\(seni\)\(forcenet.com.pl\)/\1@\2/ --- --- gg: 3973618 -------------------------------------------- ------------------------------------------------------------ -- pozdrawiam Sebastian Nieszwiec ------------------------------------------------------------ --- POLNET ------------------------------------------------ --- LAN/WAN Administrator ---------------------------------- --- email: sed -e 's/\(s\.nieszwiec\)\(polnet\.tv\)/\1@\2/ - --- gg: 3973618 -------------------------------------------- ------------------------------------------------------------

2 1

Explain a difference between filters??
by Владислав Гришин 08 Jul '10

08 Jul '10

Part of bird.conf ... filter export_default_1 { if net ~ [ 0.0.0.0/0 ] then accept; reject; } filter export_default_2 { if net ~ 0.0.0.0/0 then accept; reject; } filter export_net_1 { if net ~ [ 12.12.12.12/30 ] then accept; reject; } filter export_net_2 { if net ~ 12.12.12.12/30 then accept; reject; } bird> show route filter export_net_1 12.12.12.12/30 via 10.10.1.3 on eth1.550 [myospf 2010-05-08 18:54:23] * E2 (150/10/20) [10.10.1.3] bird> show route filter export_net_2 12.12.12.12/30 via 10.10.1.3 on eth1.550 [myospf 2010-05-08 18:54:23] * E2 (150/10/20) [10.10.1.3] bird> show route filter export_default_1 0.0.0.0/0 via 93.191.9.205 on eth2.701 [static1 2010-05-09 16:31:10] * (254) via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:54:21] (100) [AS28917i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:15] (100) [AS8732i] bird> show route filter export_default_2 0.0.0.0/0 via 93.191.9.205 on eth2.701 [static1 2010-05-09 16:31:10] * (254) via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:54:21] (100) [AS28917i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:15] (100) [AS8732i] 12.12.12.12/30 via 10.10.1.3 on eth1.550 [myospf 2010-05-08 18:54:23] * E2 (150/10/20) [10.10.1.3] 208.0.208.0/22 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:54:41] * (100) [AS26759i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:24] (100) [AS26759i] 40.0.40.0/24 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:55:07] * (100) [AS4249i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:49] (100) [AS4249i] 69.0.69.0/24 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:55:20] * (100) [AS19374i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:48] (100) [AS19374i] ... more output 8.2.0.0/24 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:55:10] * (100) [AS20473i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:25] (100) [AS20473i] 200.2.192.0/19 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:55:22] * (100) [AS10778i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:26] (100) [AS10778i] 72.2.64.0/19 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:55:08] * (100) [AS19009i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:54:28] (100) [AS19009i] 69.2.77.0/24 via 93.191.9.205 on eth2.701 [bgpfiord 2010-05-08 18:54:52] * (100) [AS15105i] via 82.138.44.129 on eth2.51 [bgpcomcor 2010-05-08 18:55:07] (100) [AS15105i] --More-- ... more output Why filters export_net_1 and export_net_2 work equally, and/but filters export_default_1 and export_default_1 on a miscellaneous? Why the filter export_default_2 "does not work" also as the filter export_net_2 works? Vladislav Grishin

2 2

Why routes have no attribute source = RTS_STATIC ?
by Владислав Гришин 08 Jul '10

08 Jul '10

Hi, ALL. Part of bird.conf : protocol static { # disabled; # Disable by default # table testable; # Connect to a non-default table preference 254; # Default preference of routes # debug { states, routes, filters, interfaces, events, packets }; # debug all; route 1.1.1.0:255.255.255.0 via "lo"; route 2.2.2.0:255.255.255.0 via "lo"; route 3.3.3.0/24 reject; route 0.0.0.0/0 via 93.191.9.205; } : Console output of birdc bird> show static static1 3.3.3.0/24 unreachable 0.0.0.0/0 via 93.191.9.205 1.1.1.0/24 dev lo 2.2.2.0/24 dev lo bird> bird> show route where source=RTS_STATIC 0.0.0.0/0 via 93.191.9.205 on eth2.701 [static1 2010-05-09 16:31:10] * (254) 3.3.3.0/24 unreachable [static1 2010-05-08 18:54:13] * (254) bird> Why 1.1.1.0/24 and 2.2.2.0/24 have no attribute source = RTS_STATIC in the bird routing table despite the fact that what 1.1.1.0/24 and 2.2.2.0/24 are configured in protocol static? Vladislav Grishin

2 2

Re: BGP newbie need help
by Ondrej Filip 04 Jul '10

04 Jul '10

On 4.7.2010 19:19, Maciej Drobniuch wrote: > 2010/7/4 Ondrej Filip<ondrej.filip(a)nic.cz>: >> On 4.7.2010 17:27, Maciej Drobniuch wrote: >>> >>> Hi All! >>> I have established a bgp session to my isp using bird. >>> I have one problem. The session works great but it doesn't advertise >>> my inetnum. I can't find it in the documentation how to define my >>> inetnum and how to advertise it? >> >> Did you define some static routes? > No, I did not. Should I create a static route for my self and > redistribute it via bgp? Yes, you should. Otherwise there is nothing to export. Which you can see from the output of show route preexport bgp1 You may add something like: protocol static { route 192.168.1.0/24 reject; } Or this could be some real route in case you do not run some IGP. Or you can export some device route if it's e.g. some anycast node etc. Ondrej >> >> What is an output of >> >> show route preexport bgp1 >> and >> show route export bgp1 >> >> ? >> birdc > BIRD 1.2.3 ready. > bird> show route preexport bgp1 > bird> show route export bgp1 > bird> > >> >> Ondrej >> > BIG Thanks! > Maciej >>> >>> My cfg: >>> protocol bgp { >>> # disabled; >>> description "lol"; >>> local as 51060; >>> neighbor 213.5.9.73 as 49895; >>> # multihop 20 via 62.168.0.13; >>> hold time 240; >>> startup hold time 240; >>> connect retry time 120; >>> keepalive time 80; # defaults to hold time / 3 >>> start delay time 5; # How long do we wait before initial >>> connect >>> error wait time 60, 300;# Minimum and maximum time we wait >>> after an error (when consecutive >>> # # errors occur, we increase the delay >>> exponentially ... >>> error forget time 300; # ... until this timeout expires) >>> # disable after error; # Disable the protocol automatically >>> when an error occurs >>> # next hop self; # Disable next hop processing and >>> always advertise our local address as nexthop >>> path metric 1; # Prefer routes with shorter paths >>> (like Cisco does) >>> default bgp_med 0; # MED value we use for comparison when >>> none is defined >>> default bgp_local_pref 0; # The same for local preference >>> source address localip; # What local address we use for >>> the TCP connection >>> # password "idonotusemd5"; # Password used for MD5 >>> authentication >>> # rr client; # I am a route reflector and the >>> neighor is my client >>> # rr cluster id 1.0.0.1; # Use this value for cluster id >>> instead of my router id >>> # export where source=RTS_STATIC; >>> export filter { >>> if source = RTS_STATIC then { >>> bgp_community = -empty-; bgp_community = >>> add(bgp_community,(51060,49895)); >>> bgp_origin = 0; >>> bgp_community = -empty-; >>> bgp_community.add((51060,49895)); >>> if (51060,49895) ~ bgp_community then >>> bgp_community.add((0, 1)); >>> if bgp_path ~ [= 51060 =] then >>> bgp_path.prepend(51060); >>> accept; >>> } >>> reject; >>> # accept; >>> }; >>> } >>> >>> Big thanks !!!! >>> >> >> >> -- >> ( CZ.NIC z.s.p.o. ) >> ------------------------------------------------- >> Ondrej Filip - CEO >> >> Office : Americka 23, Praha 2, Czech Republic >> Email : ondrej.filip(a)nic.cz http://www.nic.cz >> Private: feela(a)network.cz >> ------------------------------------------------- >> > > >

1 0

Re: BGP newbie need help
by Ondrej Filip 04 Jul '10

04 Jul '10

On 4.7.2010 17:27, Maciej Drobniuch wrote: > Hi All! > I have established a bgp session to my isp using bird. > I have one problem. The session works great but it doesn't advertise > my inetnum. I can't find it in the documentation how to define my > inetnum and how to advertise it? Did you define some static routes? What is an output of show route preexport bgp1 and show route export bgp1 ? Ondrej > > My cfg: > protocol bgp { > # disabled; > description "lol"; > local as 51060; > neighbor 213.5.9.73 as 49895; > # multihop 20 via 62.168.0.13; > hold time 240; > startup hold time 240; > connect retry time 120; > keepalive time 80; # defaults to hold time / 3 > start delay time 5; # How long do we wait before initial connect > error wait time 60, 300;# Minimum and maximum time we wait > after an error (when consecutive > # # errors occur, we increase the delay > exponentially ... > error forget time 300; # ... until this timeout expires) > # disable after error; # Disable the protocol automatically > when an error occurs > # next hop self; # Disable next hop processing and > always advertise our local address as nexthop > path metric 1; # Prefer routes with shorter paths > (like Cisco does) > default bgp_med 0; # MED value we use for comparison when > none is defined > default bgp_local_pref 0; # The same for local preference > source address localip; # What local address we use for > the TCP connection > # password "idonotusemd5"; # Password used for MD5 authentication > # rr client; # I am a route reflector and the > neighor is my client > # rr cluster id 1.0.0.1; # Use this value for cluster id > instead of my router id > # export where source=RTS_STATIC; > export filter { > if source = RTS_STATIC then { > bgp_community = -empty-; bgp_community = > add(bgp_community,(51060,49895)); > bgp_origin = 0; > bgp_community = -empty-; > bgp_community.add((51060,49895)); > if (51060,49895) ~ bgp_community then > bgp_community.add((0, 1)); > if bgp_path ~ [= 51060 =] then > bgp_path.prepend(51060); > accept; > } > reject; > # accept; > }; > } > > Big thanks !!!! >

1 0

BGP newbie need help
by Maciej Drobniuch 04 Jul '10

04 Jul '10

Hi All! I have established a bgp session to my isp using bird. I have one problem. The session works great but it doesn't advertise my inetnum. I can't find it in the documentation how to define my inetnum and how to advertise it? My cfg: protocol bgp { # disabled; description "lol"; local as 51060; neighbor 213.5.9.73 as 49895; # multihop 20 via 62.168.0.13; hold time 240; startup hold time 240; connect retry time 120; keepalive time 80; # defaults to hold time / 3 start delay time 5; # How long do we wait before initial connect error wait time 60, 300;# Minimum and maximum time we wait after an error (when consecutive # # errors occur, we increase the delay exponentially ... error forget time 300; # ... until this timeout expires) # disable after error; # Disable the protocol automatically when an error occurs # next hop self; # Disable next hop processing and always advertise our local address as nexthop path metric 1; # Prefer routes with shorter paths (like Cisco does) default bgp_med 0; # MED value we use for comparison when none is defined default bgp_local_pref 0; # The same for local preference source address localip; # What local address we use for the TCP connection # password "idonotusemd5"; # Password used for MD5 authentication # rr client; # I am a route reflector and the neighor is my client # rr cluster id 1.0.0.1; # Use this value for cluster id instead of my router id # export where source=RTS_STATIC; export filter { if source = RTS_STATIC then { bgp_community = -empty-; bgp_community = add(bgp_community,(51060,49895)); bgp_origin = 0; bgp_community = -empty-; bgp_community.add((51060,49895)); if (51060,49895) ~ bgp_community then bgp_community.add((0, 1)); if bgp_path ~ [= 51060 =] then bgp_path.prepend(51060); accept; } reject; # accept; }; } Big thanks !!!! -- Pozdrawiam! Maciej Drobniuch

1 0

bgp protocol info in passive mode
by Arjan Filius 02 Jul '10

02 Jul '10

Hello Bird-users, Currently making some monitor scripts, and working on that i noticed a little difference with 2 (ipv4 and ipv6) export-only bgp protocols. ipv4: bird> show protocols name proto table state since info <snip> bgp_Arjan_Filius BGP master start Jun25 Passive ipv6: bird> show protocols name proto table state since info <snip> bgp_Arjan_Filius BGP master start Jun25 Idle Notice the difference in the info field on both the ipv6 and ipv4 definition i'd used the "passive on;" statement Both the protocol definitions are exactly the same, apart from the neighbor definition. The only difference is the ipv4 neigbor is in a direct connected subnet, where the ipv6 isn't. After "fixing" the ipv6 passive neigbor to a direct connected ip i got the "Passive" in the info field. I'm not sure what is the correct behaviour, but displaying passive when actually passive eases my monitoring scripts. Regards -- Arjan Filius mailto:iafilius@xs4all.nl

2 2

May have found an ipa_classify bug in OSPFv3 BIRD
by Michael Adams 02 Jul '10

02 Jul '10

I was trying out BIRD on a work router, and notice that it is rejecting my ULA-based addresses on the interfaces, but not the global ones (I use two subnets: one for internal work stuff that rarely changes & one for public access that can change with our tunnel broker/v6 access). Is anyone able to enable support for the fc00::/7 range? As far as I can tell in the public GIT, the function calling the error is ipa_classify. http://en.wikipedia.org/wiki/Unique_local_address Thanks, Michael Adams http://unquietwiki.com/ The information in this email is confidential and may be legally privileged. It is intended solely for the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited and may be unlawful. The contents of this email do not necessarily represent the views or policies of E-Z Rent-A-Car or employees.

2 1