- Bird-users - Mailing Lists

bird 2.0.7 crashes with OSPF
by Nico Schottelius 14 Jan '20

14 Jan '20

Hello, continuing my travel now with OSPF, we have found two cases to crash bird via birdc: bird> show ospf topology all ospf_v4 Connection closed by server [00:50] replacement-router1.place5:~# ps aux | grep bird 14220 root 0:00 grep bird bird> show ospf state ospf_v4 Connection closed by server [23:44] router1.place6:~# birdc Unable to connect to server control socket (/var/run/bird.ctl): Connection refused The config leading to this is attached below. Best regards, Nico p.s.: In the documentation for the "instance id" it says "Default value is 0 unless OSPFv3-AF extended address families are used, see RFC 5838 for that case." And further above it says that RFC5838 is enabled by default. However shouldn't it be 64 in case of an ipv4 channel in ospf according to rfc5838? protocol ospf v3 ospf_v6 { instance id 0; ipv6 { import all; export all; }; area 0 { interface "bond0.8" { cost 5; hello 5; retransmit 2; wait 10; dead 20; }; }; } protocol ospf v3 ospf_v4 { instance id 64; ipv4 { import all; export all; }; area 0 { interface "bond0.8" { cost 5; hello 5; retransmit 2; wait 10; dead 20; }; }; } -- Modern, affordable, Swiss Virtual Machines. Visit www.datacenterlight.ch

4 5

crash in ospf lsupd/dbdes (only if authentication enabled?)
by Kenth Eriksson 14 Jan '20

14 Jan '20

We have seen a bird crash due to memory corruption. The call stacks shows that it can happen at different locations, but they all seem to come from that the packet size becomes larger than the socket tx buffer size. Before the crash happens, the following trace can be observed in the log; 2020-01-08 11:06:52.584 <DBG> Assertion '*plen < ifa->sk->tbsize' failed at /usr/local/src/tm3000/ext/bird/proto/ospf/packet.c:97 Debugging shows that *plen is 1504 (after adding auth_len of 32 bytes). The tbsize is set to 1492 which corresponds to the MTU size. The procedure ospf_pkt_maxsize does not take account for authentication, is that correct? valgrind evidence of issue; ==3272== Invalid write of size 4 ==3272== at 0xFFB8594: memcpy (in /usr/lib/valgrind/vgpreload_memcheck-ppc32-linux.so) ==3272== by 0x10015F6F: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) ==3272== by 0x10039B87: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) ==3272== by 0x1002E813: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) ==3272== by 0x1003978F: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) ==3272== by 0x10046963: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) ==3272== by 0x100473B7: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) ==3272== by 0x1000245F: ??? (in /opt/appl/cuappl04a-r34a-39/sbin/bird) (gdb) list *0x1002E813 0x1002e813 is in ospf_do_send_dbdes (/usr/local/src/builduser/rc/rel_34_0/ext/bird/proto/ospf/dbdes.c:196). 191 struct ospf_iface *ifa = n->ifa; 192 193 OSPF_PACKET(ospf_dump_dbdes, n->ldd_buffer, 194 "DBDES packet sent to nbr %R on %s", n->rid, ifa->ifname); 195 sk_set_tbuf(ifa->sk, n->ldd_buffer); 196 ospf_send_to(ifa, n->ip); 197 sk_set_tbuf(ifa->sk, NULL); 198 } 199 200 /** (gdb) list *0x10039B87 0x10039b87 is in ospf_send_to (/usr/local/src/builduser/rc/rel_34_0/ext/bird/proto/ospf/packet.c:110). 105 if (pass->alg < ALG_HMAC) 106 strncpy(auth_tail, pass->password, auth_len); 107 else 108 memset32(auth_tail, HMAC_MAGIC, auth_len / 4); 109 110 mac_fill(pass->alg, pass->password, pass->length, (byte *) pkt, *plen, auth_tail); 111 break; 112 113 default: 114 bug("Unknown authentication type"); (gdb) list *0x10015F6F 0x10015f6f is in mac_fill (/usr/local/src/builduser/rc/rel_34_0/ext/bird/lib/mac.h:112). 107 108 static inline void mac_update(struct mac_context *ctx, const byte *data, uint datalen) 109 { ctx->type->update(ctx, data, datalen); } 110 111 static inline byte *mac_final(struct mac_context *ctx) 112 { return ctx->type->final(ctx); } 113 114 static inline void mac_cleanup(struct mac_context *ctx) 115 { memset(ctx, 0, ctx->type->ctx_length); } 116 (gdb) ASAN evidence of the crash; ================================================================= ==3202==ERROR: AddressSanitizer: heap-buffer-overflow on address 0xb4105a64 at pc 0x38630004 bp 0xbf958200 sp 0xbf958218 WRITE of size 4 at 0xb4105a64 thread T0 #0 0x38630000 (+0x10630000) #1 0xfb2aa10 in __asan_report_error (/usr/lib/gcc/powerpc-unknown-linux-gnu/4.9.3/libasan.so.1+0x65a10) #2 0xfb2bd70 in __asan_report_store4 (/usr/lib/gcc/powerpc-unknown-linux-gnu/4.9.3/libasan.so.1+0x66d70) #3 0x100d26d0 in ospf_send_to (/opt/appl/cuappl04a-r34a-39/sbin/bird+0x100d26d0) 0xb4105a64 is located 0 bytes to the right of 1508-byte region [0xb4105480,0xb4105a64) allocated by thread T0 here: #0 0x10060ab4 in bird_xmalloc (/opt/appl/cuappl04a-r34a-39/sbin/bird+0x10060ab4) SUMMARY: AddressSanitizer: heap-buffer-overflow ??:0 ?? Shadow bytes around the buggy address: 0x36820af0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x36820b00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x36820b10: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x36820b20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x36820b30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 =>0x36820b40: 00 00 00 00 00 00 00 00 00 00 00 00[04]fa fa fa 0x36820b50: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa 0x36820b60: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa 0x36820b70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x36820b80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x36820b90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Shadow byte legend (one shadow byte represents 8 application bytes): Addressable: 00 Partially addressable: 01 02 03 04 05 06 07 Heap left redzone: fa Heap right redzone: fb Freed heap region: fd Stack left redzone: f1 Stack mid redzone: f2 Stack right redzone: f3 Stack partial redzone: f4 Stack after return: f5 Stack use after scope: f8 Global redzone: f9 Global init order: f6 Poisoned by user: f7 Contiguous container OOB:fc ASan internal: fe ==3202==ABORTING

3 7

Bird exporter
by Irene Lalioti 13 Jan '20

13 Jan '20

Hello guys! I would like to ask you, if anyone of you has deployed bird_exporter for Prometheus metrics to Grafana dashboard? https://github.com/czerwonk/bird_exporter I have setup Grafana dashboards to visualize the metrics Prometheus collects. I have imported the Dashboard for the bird from github. Now I am not sure how to orchestrate the bird_exporter into the bird config, other than : 1. timeformat protocol iso long; 2. bird_protocol_prefix_import_count{name="bgp1",proto="BGP",ip_version="4"} 600000 bird_protocol_prefix_import_count{name="bgp1",proto="BGP",ip_version="6"} 50000 bird_ospfv3_running{name="ospf1"} 1 as instructed in the doc. My bird is 2.0.7. thanks a lot in advance! Irene || || || || || -- Irene Lalioti Network Engineer Fondation RESTENA 2, avenue de l'Université L-4365 Esch/Alzette Tel: +352 424409 1 Fax: +352 422473

1 0

RPKI doesn’t work with a FQDN
by Alarig Le Lay 11 Jan '20

11 Jan '20

Hi, I have this configuration: asbr02 ~ # cat /etc/bird.conf.d/protocol_rpki/* protocol rpki rpki_alarig { roa4 { table r4; }; roa6 { table r6; }; #remote "msi.no.swordarmor.fr"; remote 2a0e:f42::1; } protocol rpki rpki_conan { roa4 { table r4; }; roa6 { table r6; }; remote "conan.grifon.fr"; } The first protocol establish with the second fails bird> show protocols all rpki_alarig Name Proto Table State Since Info rpki_alarig RPKI --- up 15:54:25.902 Established Cache server: 2a0e:f42::1:323 Status: Established Transport: Unprotected over TCP Protocol version: 1 Session ID: 28569 Serial number: 285 Last update: before 84.055 s Refresh timer : 323.944/408 Retry timer : --- Expire timer : 7115.944/7200 Channel roa4 State: UP Table: r4 Preference: 100 Input filter: ACCEPT Output filter: REJECT Routes: 100516 imported, 0 exported, 100516 preferred Route change stats: received rejected filtered ignored accepted Import updates: 100816 0 0 0 100816 Import withdraws: 300 0 --- 0 300 Export updates: 0 0 0 --- 0 Export withdraws: 0 --- --- --- 0 Channel roa6 State: UP Table: r6 Preference: 100 Input filter: ACCEPT Output filter: REJECT Routes: 16578 imported, 0 exported, 16578 preferred Route change stats: received rejected filtered ignored accepted Import updates: 16641 0 0 0 16641 Import withdraws: 63 0 --- 0 63 Export updates: 0 0 0 --- 0 Export withdraws: 0 --- --- --- 0 bird> show protocols all rpki_conan Name Proto Table State Since Info rpki_conan RPKI --- start 15:54:25.847 Transport-Error Cache server: conan.grifon.fr:323 Status: Transport-Error Transport: Unprotected over TCP Protocol version: 1 Session ID: --- Serial number: --- Last update: --- Refresh timer : --- Retry timer : 451.669/600 Expire timer : --- Channel roa4 State: DOWN Table: r4 Preference: 100 Input filter: ACCEPT Output filter: REJECT Channel roa6 State: DOWN Table: r6 Preference: 100 Input filter: ACCEPT Output filter: REJECT I see the DNS request (and the answer): 15:54:25.851095 IP6 asbr02.cogent-rns.grifon.fr.35411 > drogon.grifon.fr.domain: 167+ A? conan.grifon.fr. (33) 15:54:25.851105 IP6 asbr02.cogent-rns.grifon.fr.35411 > drogon.grifon.fr.domain: 14516+ AAAA? conan.grifon.fr. (33) 15:54:25.851495 IP6 drogon.grifon.fr.domain > asbr02.cogent-rns.grifon.fr.35411: 167 1/0/0 A 89.234.186.8 (49) 15:54:25.851515 IP6 drogon.grifon.fr.domain > asbr02.cogent-rns.grifon.fr.35411: 14516 1/0/0 AAAA 2a00:5884::8 (61) But no SYN over 323. However, I can telnet to it: asbr02 ~ # mtr -bzwe msi.no.swordarmor.fr Start: Sat Jan 11 15:55:59 2020 HOST: asbr02.cogent-rns.grifon.fr Loss% Snt Last Avg Best Wrst StDev 1. AS204092 regis.swordarmor.fr (2a00:5884::1f) 0.0% 10 0.2 0.2 0.1 0.3 0.0 2. AS208627 tinc0.core02-arendal.no.swordarmor.fr (2a0e:f42:fffe::6) 0.0% 10 51.5 51.7 51.2 52.2 0.0 3. AS208627 msi.no.swordarmor.fr (2a0e:f42::1) 0.0% 10 52.0 52.3 51.4 52.8 0.0 asbr02 ~ # mtr -bzwe conan.grifon.fr Start: Sat Jan 11 15:57:47 2020 HOST: asbr02.cogent-rns.grifon.fr Loss% Snt Last Avg Best Wrst StDev 1. AS204092 conan.grifon.fr (2a00:5884::8) 0.0% 10 0.3 0.3 0.2 0.5 0.0 asbr02 ~ # telnet msi.no.swordarmor.fr 323 Trying 2a0e:f42::1... Connected to msi.no.swordarmor.fr. Escape character is '^]'. ^] telnet> quit Connection closed. asbr02 ~ # telnet conan.grifon.fr 323 Trying 2a00:5884::8... Connected to conan.grifon.fr. Escape character is '^]'. ^] telnet> quit Connection closed. And then I see the SYN: 16:01:28.787297 IP6 asbr02.cogent-rns.grifon.fr.60330 > conan.grifon.fr.323: Flags [S], seq 1340260165, win 28800, options [mss 1440,sackOK,TS val 4034128416 ecr 0,nop,wscale 7], le ngth 0 16:01:28.787677 IP6 conan.grifon.fr.323 > asbr02.cogent-rns.grifon.fr.60330: Flags [S.], seq 287295091, ack 1340260166, win 64260, options [mss 1440,sackOK,TS val 4292064010 ecr 403 4128416,nop,wscale 7], length 0 16:01:28.787713 IP6 asbr02.cogent-rns.grifon.fr.60330 > conan.grifon.fr.323: Flags [.], ack 1, win 225, options [nop,nop,TS val 4034128416 ecr 4292064010], length 0 16:01:31.114241 IP6 asbr02.cogent-rns.grifon.fr.60330 > conan.grifon.fr.323: Flags [F.], seq 1, ack 1, win 225, options [nop,nop,TS val 4034130743 ecr 4292064010], length 0 16:01:31.114709 IP6 conan.grifon.fr.323 > asbr02.cogent-rns.grifon.fr.60330: Flags [F.], seq 1, ack 2, win 503, options [nop,nop,TS val 4292066337 ecr 4034130743], length 0 16:01:31.114725 IP6 asbr02.cogent-rns.grifon.fr.60330 > conan.grifon.fr.323: Flags [.], ack 2, win 225, options [nop,nop,TS val 4034130743 ecr 4292066337], length 0 The first protocol only established when I put the IP address directly. Plus, not having the brackets over the literal IPv6 address is a bit confusing. The IP isn’t 2a0e:f42::1:323. Regards, -- Alarig

2 2

Multi thread support
by Marcelo Balbinot 10 Jan '20

10 Jan '20

Hello, About multi thread support (cpu multi-core use) is it in the roadmap? thanks -- Marcelo Balbinot IP/MPLS Backbone Manager at AS53062 G G NET Telecomunicações www.gegnet.com.br INOC 53062*100

2 1

[RFC PATCH 00/20] Context typed filters
by Asbjørn Sloth Tønnesen 08 Jan '20

08 Jan '20

This RFC patch series is a first stab, at adding some syntactic sugar making it more plessant to implement non-trivial dual-stack configurations, like https://bgpfilterguide.nlnog.net/ I'm not completely happy with the naming at this stage, but I want to get this series out before 36C3 ends. Looking forward to your feedback. Asbjørn Sloth Tønnesen (20): add NET_ANY make symbol types maskable add new symbol flags for typed symbols add support for defining typed symbols show symbols: don't show typed symbols add initial FI_TYPED_CONSTANT instruction add generic pseudo filter check function add f_has_typed_symbols helper function use net_type when looking up symbols in SYM_FLAG_SAME check show symbol flags in dump filter all check if functions need to be adapted force consistent number of function arguments use FI_TYPED_CONSTANT for typed constants adapt infrastructure adapt protocol filters adapt sh route filters adapt mrt filters allow typed definitions add typed test config add typed documentation conf/cf-lex.l | 75 ++++++++++++++++++++---- conf/conf.h | 58 +++++++++++++----- conf/confbase.Y | 6 +- doc/bird.sgml | 10 ++++ filter/Makefile | 2 +- filter/adapt.c | 150 +++++++++++++++++++++++++++++++++++++++++++++++ filter/config.Y | 28 +++++++-- filter/decl.m4 | 77 +++++++++++++++++++++--- filter/f-inst.c | 42 ++++++++++++- filter/filter.c | 29 +++++---- filter/filter.h | 16 +++++ filter/test.conf | 41 +++++++++++++ lib/net.c | 1 + lib/net.h | 1 + nest/cmds.c | 3 + nest/config.Y | 8 ++- nest/route.h | 1 + nest/rt-show.c | 9 ++- proto/mrt/mrt.c | 17 +++++- proto/mrt/mrt.h | 2 + 20 files changed, 519 insertions(+), 57 deletions(-) create mode 100644 filter/adapt.c -- 2.24.0

2 3

Static route tracking with BFD
by Юрий Иванов 08 Jan '20

08 Jan '20

Hi, I've successfully bring bfd session up with my router. All seems good, but in some cases I will need to flush static route if bfd session comes down. As far as I know BIRD 2.0.7 support this but how to configure it?

2 3

Fwd: [PDB Announce] Call for volunteers for PeeringDB
by Chriztoffer Hansen 07 Jan '20

07 Jan '20

Fwd to bird-users list, because I thought mayhaps (individual_count >= 1) might be interested... Best Regards, Chriztoffer ---------- Forwarded message --------- From: Stephen McManus <smcmanus(a)peeringdb.com> Date: Mon, 6 Jan 2020 at 22:10 Subject: [PDB Announce] Call for volunteers for PeeringDB To: <pdb-announce(a)lists.peeringdb.com> Hi all, We are looking for volunteers to serve on the PeeringDB Admin and Product Committees. The ideal volunteers are industry professionals who are passionate about improving PeeringDB, and ensuring it meets the needs of the community it supports. We are most interested in people who have significant free time to dedicate to these efforts (as much as 4 or more hours a week). Please don’t volunteer if you can’t commit the time! Admin Committee: • Review requests from community members • Authorize users to organizations and networks • Generally keep PeeringDB’s data trustworthy and correct Product Committee: • Review and prioritize requests for improvements to PeeringDB • Work with developers and community members to specify how to solve problems and implement improvements • Set vision for how PeeringDB can help the community in the future If you are interested in either of these roles, and have time you can commit, please let us know at stefan(a)peeringdb.com for the Admin Committee or smcmanus(a)peeringdb.com for the Product Committee. Please include a sentence or two about yourself, why you want to volunteer and how you think you can help. Thanks! -Steve _______________________________________________ Pdb-announce mailing list Pdb-announce(a)lists.peeringdb.com https://lists.peeringdb.com/cgi-bin/mailman/listinfo/pdb-announce

1 0

Follow up question on "next hop self"
by Nico Schottelius 07 Jan '20

07 Jan '20

Hello, I was wondering about how bird solves the following problem with next hop self enabled: R1 <--------> R2 \ / \ / upstream 1 So both routers receive the same routes from upstream and insert them with the same local_pref. With "next hop self" on, both routers will think that the path is not the same and potentially either of the two routers could start routing via the other router instead of via upstream directly. What is the typical approach to solve this situation correctly? Does making setting the local_pref lower if received from an iBGP router make sense in that case? Best regards, Nico -- Modern, affordable, Swiss Virtual Machines. Visit www.datacenterlight.ch

4 8

Problem with if statements containing && or || expressions
by mikma.bird＠lists.m7n.se 07 Jan '20

07 Jan '20

Hello, I'm having problems with if statements containing &&, or ||. Why do the config file below cause the following error? $ ./bird -d -l bird: filters, line 8: Argument 2 of AND must be of type bool, got type int bird: bird.conf:15:12 Runtime error while evaluating expression; see log for details # bird.conf function dummy() { return 1; } function test() { if (dummy() != 0) && true then { print "Success"; } return 0; } eval test(); protocol device { interface "*"; } /Mikael

2 4