- Bird-users - Mailing Lists

Gateway on routed network problem (aka: Received route with strange next-hop)
by Alexander Demenshin 04 Sep '14

04 Sep '14

Hi, I've a static protocol which has routes like: route 10.0.0.0/24 via "uplink"; route 192.168.0.0/24 via 10.0.0.1; Obviously, this does not work in Bird, though this is perfectly valid and accepted by the Linux kernel (as long as "uplink" is up). When adding this route manually (ip route), I see in the log file messages like: KRT: Received route 192.168.0.0/24 with strange next-hop 10.0.0.1 I've found why this is so (from http://permalink.gmane.org/gmane.network.bird.user/2042) "BIRD generally assumes tha that gateway of routes is a direct neighbor according to IP addresses on interfaces, not according to routing tables." Well, now the question - *why* Bird assumes so and how to remove this assumption? In my situation, "uplink" is actually ethernet, not any kind of PtP interface, and I really want to avoid adding local IP which is in gateway's network (in some cases this is simply impossible due to tiny transfer networks and multiple gateways). Is there any easy fix for this? I feel that only few lines could be changed to allow this, but I don't know the code well (yet). Thank you! -- With best regards, Alexander.

2 2

BFD implementation in 1.4.0
by Aleksey Berezin 02 Sep '14

02 Sep '14

Hi there! Firstly, sorry for my French ;) Recently I tried to test BFD implementation in 1.4.0 BIRD release. Case is "OSPF ECMP with less-second failover". On one side - Extreme Networks switches, other side - several boxes with Debian and BIRD, of course. So, I find out several annoying things. First thing: BIRD BFD implementation uses non-RFC5881 source UDP port for outgoing packets. Here some tcpdump: =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 20:34:21.355747 <BIRD_BOX_MAC> > <EXTREME_SW_MAC>, ethertype 802.1Q (0x8100), length 70: vlan <vlan>, p 0, ethertype IPv4, (tos 0xc0, ttl 255, id 20806, offset 0, flags [none], proto UDP (17), length 52) <BIRD_BOX_IP>.39156 > <EXTREME_SW_IP>.3784: BFDv1, length: 24 Control, State Init, Flags: [none], Diagnostic: No Diagnostic (0x00) Detection Timer Multiplier: 5 (5000 ms Detection time), BFD Length: 24 My Discriminator: 0x8a82eb8a, Your Discriminator: 0x00000002 Desired min Tx Interval: 1000 ms Required min Rx Interval: 10 ms Required min Echo Interval: 0 ms =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= I saw many messages in switch logs: > <Warn:BFD.RxPktInv> Received BFD an invalid packet on VLAN <vlan> Error:Invalid UDP src port Number BFD session between BIRD and Extreme Networks switch can't be established. So, src port for outgoing BFD-packet from BIRD is 39156, when RFC5881 contains: =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 4. Encapsulation BFD Control packets MUST be transmitted in UDP packets with destination port 3784, within an IPv4 or IPv6 packet. The source port MUST be in the range 49152 through 65535. =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Ok, fixing (only for test purposes): =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= root@debian-1:~# sysctl net.ipv4.ip_local_port_range net.ipv4.ip_local_port_range = 32768 61000 root@debian-1:~# sysctl -w net.ipv4.ip_local_port_range="49152 65535" net.ipv4.ip_local_port_range = 49152 65535 root@debian-1:~# sysctl net.ipv4.ip_local_port_range net.ipv4.ip_local_port_range = 49152 65535 =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= After this BFD session was estabilished. Of course, net.ipv4.ip_local_port_range tuning seems to be not usable in production environment, and need to be fixed in BIRD, I think. Just because some network hardware vendors interprets RFC's exactly as written ;) But now - second thing: 'protocol bfd' block in bird.conf useless. Here part of config: =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= protocol bfd FOR_OSPF { interface <BIRD_BOX_IP>/32 { interval 50 ms; min rx interval 50 ms; # presence of this string by the docs does not matter min tx interval 50 ms; # presence of this string by the docs does not matter multiplier 3; }; neighbor <EXTREME_SW_IP> dev "<BIRD_BOX_IP>" local <BIRD_BOX_IP>; # presence of this string by the docs does not matter }; protocol ospf OSPF1 { ... area 0.0.0.0 { interface <BIRD_BOX_IP>/32 { ... bfd; }; }; ... }; =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= And what I see with tcpdump: =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 13:09:05.433336 <BIRD_BOX_MAC> > <EXTREME_SW_MAC>, ethertype 802.1Q (0x8100), length 70: vlan <vlan>, p 0, ethertype IPv4, (tos 0xc0, ttl 255, id 6662, offset 0, flags [none], proto UDP (17), length 52) <BIRD_BOX_IP>.62698 > <EXTREME_SW_IP>.3784: BFDv1, length: 24 Control, State Up, Flags: [none], Diagnostic: No Diagnostic (0x00) Detection Timer Multiplier: 5 (500 ms Detection time), BFD Length: 24 My Discriminator: 0x35e75227, Your Discriminator: 0x00000007 Desired min Tx Interval: 100 ms Required min Rx Interval: 10 ms Required min Echo Interval: 0 ms =-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= There is no any configured intervals and multiplier. Just default values. I think this is not correct. Maybe I misconfigure BIRD? Please advise. Thanks for attention.

4 13

export filter that checks for existance of another route?
by Paul J R 02 Sep '14

02 Sep '14

Hi All, First time post, i've been playing with bird for some time, but there is one thing i cant figure out how to do and that is use the existence of a specific route as part of a filter. for example, if the filter is processing a route to 4.3.2.0/24, check if a route to 1.2.3.4/32 (exact) exists, if so then accept, otherwise reject. Is this something that is possible? or do route filters only have the ability to check against the information thats sent to the filter itself? Thanks in advance, PJR.

3 3

Limit on kernel table number
by Alexander Demenshin 02 Sep '14

02 Sep '14

Hi, It seems that there is a limit on maximum kernel table number (in kernel protocol), in particular I couldn't use number 1000 while this is perfectly acceptable by the Linux kernel. Is there any specific reason for this limitation? Thank you! -- With best regards, Alexander.

3 3

show route all filter ...
by Mikhail A. Grishin 02 Sep '14

02 Sep '14

Hi, We use special command (see below) to view information regarding prefixes containing particular BGP communities. When such commands applied, we noticed in logs ---- Aug 27 11:50:17 ekt-rsm1 bird: Filter <NULL> did not return accept nor reject. Make up your mind Aug 27 11:50:17 ekt-rsm1 last message repeated 2690 times ---- Could you explain why? Is there any way to avoid such messages and stay at current logging level? Aug 27 11:50:17 ekt-rsm1 bird: CLI connect Aug 27 11:50:17 ekt-rsm1 bird: CLI: show route all filter {if (0,48642) ~ bgp_community then accept;} table master Aug 27 11:50:17 ekt-rsm1 bird: Filter <NULL> did not return accept nor reject. Make up your mind Aug 27 11:50:17 ekt-rsm1 last message repeated 2690 times Aug 27 11:50:17 ekt-rsm1 bird: CLI connection closed -- Best regards, Mikhail A. Grishin <m.grishin(a)msk-ix.ru>

4 3

Multiple routing tables with multiple gateways
by Fernando Frediani 01 Sep '14

01 Sep '14

Hello there, I'm currently writing a project document which consists in a server to be hosted in a Internet Exchange Point, (similar to a route server but with another propose) and one of the things it must do is: When it receives a request coming from an IP range from ISP A it must go out to the internet using the ISP A gateway, when the request comes from ISP B IP range it goes out via that ISP router and so on. Any request originating on the server itself goes out via the Internet Exchange connectivity. I know this can be done using IProute2 and creating multiple routing tables, but I would rather automate it using BIRD, so the ISPs would establish a BGP session to the server, announce their IP range, but in a different way they do with a Route server as they would need also to announce default route, and BIRD would take care to inform the kernel for when the request comes from those IPs it needs to use the specific ISP gateway for that. Would anyone be able to point me to the right direction in terms of BIRD BGP config in order to achieve that, specially with regards the main differences to a Route Server. Thanks Best regards, Fernando

1 0

OSPF Route
by Jason Aeschilman 26 Aug '14

26 Aug '14

I am using bird 1.4.0. I configured OSPF and the routes that bird adds to Linux have a metric of 0. Then I added a static route with the same network and mask and it has a metric of 1. So now a route that OSPF has added has preference over a static route but this is not the behavior I see with quagga or Cisco. With them, the static route has preference. quagga adds routes with metrics (values 11+) relative to OSPF cost while Cisco does not add an OSPF route to the FIB if a matching static route already exists. My expectation, if bird is going to use metric 0, is that it should not add an OSPF route if a static route already exists with the same network and mask. Or if it must add an OSPF route, it should have a metric such that the static route takes precedence. Is it possible to change this behavior in the config? If so, how?

1 0

Log rotation support
by Alexander Demenshin 26 Aug '14

26 Aug '14

Hi, It looks like that there is no log rotation possible when logging to files. Though I can fix it locally, it would be nice to include support for this in the future versions, as log files may be quite large and not everyone is using syslog. It could be done using birdc or some (unused yet) signal. Thank you! -- With best regards, Alexander.

3 4

Reloading static/direct/device/kernel protocols via birdc fails
by Alexander Demenshin 22 Aug '14

22 Aug '14

Hi, When I am attempting to reload kernel, static, direct or device protocol via birdc, I always get message like this "krt_main: reload failed", while in the log file I see just reloading message: <INFO> Reloading protocol krt_main For pipe, bgp and ospf instances there is only "reloading" message. Is this a bug, feature or something else? PS: I need reload to force bird to rescan specific kernel table. Thank you! -- With best regards, Alexander.

2 2

changes somewhere between 1.3.9 and 1.4.0
by Mikhail A. Grishin 22 Aug '14

22 Aug '14

Hi, We found that in 1.4.0 and later versions our approach to see rejected BGP routes for IXP customers stop working. This worked fine in 1.3.9 and earlier versions: --------- [18:44:27]ekt-rsm1% echo 'show route protocol R2854x1' | birdc | wc -l 238 [15:58:45]ekt-rsm1% echo 'show route protocol R2854x1 table T2854x1' | birdc | wc -l 245 ---------- Diffs between this two outputs - rejected routes. At 1.4.0 and later we see the same numbers: [16:00:26]ekt-rsm2# echo 'show route protocol R2854x1 table T2854x1' | birdc | wc -l 245 [16:08:51]ekt-rsm2# echo 'show route protocol R2854x1' | birdc | wc -l 245 Can you tell us, what changed? And how to restore old behavior? -- Wbr, Mikhail MSK-IX

2 1