- Bird-users - Mailing Lists

RFC 5575 feasibility check
by Alexander Shikov 23 Sep '21

23 Sep '21

Hello! RFC5575 and newer RFC8955 define validation procedure for flowspec routes received by BGP protocol. While testing flowspec in bird 2.0.8 I've found strange behavior: # birdc show route protocol BGP4_IPPORT filtered BIRD 2.0.8 ready. Table T4_IPPORT: 185.23.164.48/28 unicast [BGP4_IPPORT 18:02:30.148] * (100) [AS8647i] via 193.25.180.134 on bxe0 -- the unicast route is filtered by import policy. # birdc show route BIRD 2.0.8 ready. Table FLOW4: flow4 { dst 185.23.164.48/28; proto 1; icmp type 8; } [BGP4_IPPORT 18:02:31.214 from 193.25.180.134] * (100) [AS8647i] -- flow route remains. So the question, is feasibility check implemented in bird? Thank you! -- Alexander Shikov Technical Staff, Digital Telecom IX Tel.: +380 44 201 14 07 Mob.: +380 50 410 30 57 http://dtel-ix.net/

2 1

latest bgp perf tests blog
by Justin Pietsch 22 Sep '21

22 Sep '21

Hi, my latest blog post on bgp perf testing: https://elegantnetwork.github.io/posts/bird-on-bird-bgp-perf-episode4/. I used the code that Maria added to use BIRD as a tester, and it works better, but shows some issues with BIRD as the target. Justin

1 0

Mitigations/tunables for reducing netlink loss?
by Dave Johnson 21 Sep '21

21 Sep '21

Hi, I'm seeing these warnings quite frequently on a system that has full-internet table programmed into Linux kernel. Sep 20 11:50:48 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:50:57 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:50:59 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:00 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:01 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:05 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:05 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:07 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:10 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:10 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:14 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:17 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:19 ganges bird: Kernel dropped some netlink messages, will resync on next scan. Sep 20 11:51:19 ganges bird: Kernel dropped some netlink messages, will resync on next scan. First off, I'm not expecting wonders here, I don't expect having near to 1m routes in Linux to work flawlessly however just looking to improve on the situation. My setup: bird 1.6.6 (Debian 10) with Linux 4.9.xxx Internet routes are segregated into a dedicated kernel routing table (#99 below), no non-bird routes are in this table. prot kernel is export only (again, no non-bird routes in this kernel table) bird table "internet" contains a protocol kernel, a protocol pipe, and 3 protocol bgps (for 3 iBGP peers). prot kernel config is: protocol kernel kernel_internet { debug { states, interfaces }; table internet; kernel table 99; scan time 60; persist; learn off; graceful restart on; import none; export filter { if net ~ IP_MY_NET_PLUS then reject; if net ~ IP_CORE_NET then reject; accept; }; } I'm seeing netlink drops when upstream internet churn is say more than 200 updates/sec or so, not huge, but quite freqent and can continue for minutes/hours. Some items I've investigated so far: Increasing net.core.rmem_max and net.core.wmem_max sysctls doesn't seem to help much, strace of bird doesn't indicate any EAGAIN or blocking when writing to the netlink sockets. strace shows some room for optimization in the prot kernel (these would obviously be code changes). For example, when a route changes next-hop/interface, 2 netlink messages are sent, delete followed by add, instead of a single change/replace (this would complicate bird, but reduce netlink message in half for updates). There is plenty of cpu cycles available, bird is <%1, etc... Any pointers on tuning or config changes that may help here are appreciated. -- Dave

4 6

Bird 2 config syntax
by Robert Sander 21 Sep '21

21 Sep '21

Hi, why is it not possible to define BGP sessions like this? protocol bgp g60 { local as 64501; neighbor as 64499; ipv4 { import all; export all; }; local 192.0.2.4; neighbor 192.0.2.5; }; ipv6 { import all; export all; local 2001:db8::4; neighbor 2001:db8::5; }; }; It seems still necessary to write two bgp definitions separately for IPv4 and IPv6 repeating the AS numbers and name. I see no benefit from the Bird 1.6 notation here. Regards -- Robert Sander Heinlein Consulting GmbH Schwedter Str. 8/9b, 10119 Berlin http://www.heinlein-support.de Tel: 030 / 405051-43 Fax: 030 / 405051-19 Zwangsangaben lt. §35a GmbHG: HRB 220009 B / Amtsgericht Berlin-Charlottenburg, Geschäftsführer: Peer Heinlein -- Sitz: Berlin

3 2

Show all ROAs for an ASN
by Darren O'Connor 20 Sep '21

20 Sep '21

I have an rpki-rtr session to my bird instance. All v4 roas go to a table called roa_v4, and v6 goes to roa_v6 - Is there a way in a single query to show me all ROAs from an ASN without listing them all? Right now I'm doing something like this, which is extremely messy: birdc 'show route all table roa_v4' | grep AS13414 It would be ideal to have something like show route all table roa_v4 source 13414 - Am I missing something here? Thanks Darren

2 2

Compilation directive LOCAL_DEBUG and DBG()
by Senthil Kumar Nagappan 17 Sep '21

17 Sep '21

Hi, To enable debug outputs using DBG macro, it requires LOCAL_DEBUG or GLOBAL_DEBUG to be defined during compilation. 1. Defined it in the Makefile as below. Is there an option without changing the auto generated Makefile? CFLAGS=$(CPPFLAGS) -g -O2 -pthread -fno-strict-aliasing -fno-strict-overflow -flto -Wall -Wextra -Wstrict-prototypes -Wno-parentheses -Wno-pointer-sign -Wno-missing-field-initializers -DLOCAL_DEBUG 2. If we try to define LOCAL_DEBUG in the code, there are undef directive in the different places of the code, so may not work consistently. 3. If we define this LOCAL_DEBUG or GLOBAL_DEBUG, there are compilation errors that needs to be addressed. Thanks and Regards,Senthil

2 4

OSPF route in database not imported in table
by Joachim Desroches 16 Sep '21

16 Sep '21

Ave, I am observing behavior that I do not understand on two of our bird instances, maybe you guys can help me make sense of it. I'll start by giving you an overview of our setup, then I'll describe the problem and paste some configuration snippets. We have three pairs of routers connected over a meshed VPN, each pair being in a different physical location. These routers are connected to the wider internet, to the inter-location mesh VPN and to stub networks to whom they provide connectivity. We use BGP sessions at each site with our local upstream to obtain and export routes, and OSPF over the mesh VPN for internal routing. Everything is dual-stacked, but the issue I am observing is only over IPv4, so let us only consider IPv4. Calling the three locations A, B and C, I have the following problem: C exports a /30 prefix via OSPF (gathered by the automagic generation of stub networks). A and B correctly recieve it and display it in the `birdc show ospf state <ospf ipv4 name>` map, but while it is correctly imported into the master4 table in B, it does not appear in the master4 table in A. Both configurations are, as far as I can tell, identical for all relevant purposes. Versions of bird, operating system and kernel are the same. Relevant birdc output: ``` show ospf state ospf4: [...] router 0.0.2.1 distance 10 network [0.0.2.1-6] metric 10 stubnet xx.xxx.xxx.x/30 metric 10 ``` ``` show route table master4 protocol ospf4: Table master4: [...] ``` where the second output does not contain the required /30. Relevant configuration: ``` protocol ospf v3 ospf4 { description "backbone OSPF v4"; instance id 64; ipv4 { import all; export none; }; area 0 { interface "tun0"; interface "vlan10"; }; } ``` The main difference between A and B are the routes that they recieve over BGP from our upstreams: one of them contains a 0.0.0.0/0 default route (A) and the other does not, only the list of IPv4 routes. I am at loss as to why the route is not imported into the master table, so any advice or suggestions would be very welcome. Vale, -- Joachim Desroches

1 0

BGP "import table"
by Lutz Preßler 09 Sep '21

09 Sep '21

Hello, https://bird.network.cz/?get_doc&v=20&f=bird-6.html#bgp-import-table states import table switch A BGP import table contains all received routes from given BGP neighbor, before application of import filters. It is also called Adj-RIB-In in BGP terminology. BIRD BGP by default operates without import tables, in which case received routes are just processed by import filters, accepted ones are stored in the master table, and the rest is forgotten. Enabling import table allows to store unprocessed routes, which can be examined later by show route, and can be used to reconfigure import filters without full route refresh. Default: off. (and following "export table"). Haven't checked the filter reconfiguration/rpki_reload aspect of the feature or looked into that part of the code yet. But I cannot find a way with 2.0.8 to "examime by show route unprocessed routes". What do I miss? Regards, Lutz Preßler (SerNet GmbH)

2 3

Next bird release timeframe
by Etienne Champetier 08 Sep '21

08 Sep '21

Hello, Is there a rough estimate of when the next release of bird will be cut ? (I'm interested in the OSPF + /32 fixes) Thanks Etienne

1 0

default protocol preference values
by Kseniya Blashchuk 07 Sep '21

07 Sep '21

Hello! I could not find any information about default protocol preference values (administrative distance) for bird. Can anybody please help with it? Are there any default values for OSFP, BGP, static, etc.? Or is it better to configure each value explicitly?

3 3