- Bird-users - Mailing Lists

Re: Segmentation fault on EOR
by Ondrej Zajicek 20 Jun '14

20 Jun '14

On Fri, Jun 20, 2014 at 08:32:29PM +0400, Evgeny Uskov wrote: > Hello, > > No, the bug I'm talking about was not fixed. It can be easily > reproduced on any of the versions 1.4.1, 1.4.2, 1.4.3 (by sending the > UPDATE message from my previous letter). You are right, i conflated two different bugs. This one is already fixed but only in git: https://gitlab.labs.nic.cz/labs/bird/commit/66370eac1f88ec74b3017f1c39038d7… > In fact, this bug is not related to IPv6. It may appear even if bird > is used only for IPv4 routing, because multiprotocol extensions may be > also used in this case. Well, BIRD does not support IPv4 NLRI in multiprotocol extensions. -- 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."

1 0

Segmentation fault on EOR
by Evgeny Uskov 20 Jun '14

20 Jun '14

Hello, bird with version >= 1.4.1 has the following bug: it falls with segmentation fault when it receives certain BGP UPDATE messages. In particular, the problem is triggered by a message with attributes but without NLRI section. An example of such message is End-of-RIB marker (an empty UPDATE) with additional multiprotocol extension attributes (such as MP_UNREACH_NLRI): 0000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0010 00 1e 02 00 00 00 07 90 0f 00 03 00 01 01 The problem is caused by null pointer reference in the function bgp_set_next_hop: struct eattr *nh = ea_find(a->eattrs, EA_CODE(EAP_BGP, BA_NEXT_HOP)); ip_addr *nexthop = (ip_addr *) nh->u.ptr->data; If the case of the UPDATE message above, a->eattrs field is NULL, and hence nh is NULL too, resulting in SIGSEGV. In bird <= 1.4.0 this function is called in bgp_do_rx_update only when nlri_len != 0 (I'm not sure it's safe enough, but at least it doesn't cause this problem). However, in v1.4.1 this behavior has been changed, and now bgp_set_next_hop is called for every UPDATE message with positive attr_len + nlri_len.

2 1

about lo:0 interface
by 曾小小 20 Jun '14

20 Jun '14

hi My system is Ubuntu 13.04, kernel version is 3.8.0-21-generic, bird version is 1.3.10. I want to achieve lo: 0 (eg: 192.168.1.1/32) address as the route-id of the device, while learning through ospf routing protocol ip address to manage the device through this ip address. Configured as follows: 1: I'm on a device configured lo:0 interface address: ifconfig lo:0 192.168.1.1/32 root @ sr1 :/ usr / local / bird # ifconfig lo:0 lo: 0 Link encap: Local Loopback inet addr: 192.168.1.1 Mask: 0.0.0.0 UP LOOPBACK RUNNING MTU: 65536 Metric: 1 2: The bird in the configuration is as follows: In direct proto configured lo: 0 Interface protocol direct { interface "eth1", 192.168.1.1/32; } In the kernel configuration is as follows: protocol kernel { # Learn; # Learn all alien routes from the kernel # Persist; # Don't remove routes on bird shutdown device routes; scan time 15; # Scan kernel routing table every 15 seconds import none; # Default is import all export all; # Default is export none # Kernel table 5; # Kernel table to synchronize with (default: main) } The problem is found in bird default route pointing to the interface, and the interface address lo not found in the routing table, and why? bird> show interfaces 1001-lo up (index = 1) 1004 - MultiAccess AdminUp LinkUp Loopback Ignored MTU = 65536 1003 - 192.168.1.1 / 0 (Primary, scope univ) 127.0.0.1 / 8 (Unselected, scope host) root @ sr1 :/ usr / local / bird # birdc-v 0001 BIRD 1.3.10 ready. bird> show route 1007-0.0.0.0 / 0 dev lo [direct1 16:55] * (240) thanks.

2 1

route attribute scope
by Frederik Kriewitz 20 Jun '14

20 Jun '14

Hello, we noticed something which feels like an inconsistency/bug to us. So far we assumed that route attributes can be set in any filter of any protocol. But it seems like OSPF attributes can be set only in export filters of OSPF protocols. Here's an example configuration (Tested with BIRD 1.4.3): protocol static s_attr_test { import filter { bgp_community.add((65535,65281)); # working bgp_local_pref = 99; # working ospf_metric1=11; # not working ospf_metric2=12; # not working ospf_tag=123; # not working accept; }; route 10.50.0.0/30 via "lo"; } This is the result: # birdc show route all protocol s_attr_test BIRD 1.4.3 ready. 10.50.0.0/30 dev lo [s_attr_test 2014-06-20 12:35:25] * (200) Type: static-device unicast univ BGP.local_pref: 99 BGP.community: (65535,65281) As you can see the BGP attributes have been added correctly but the OSPF attributes are missing. Before we look further into this we would like to know if this is the intended behavior. Best Regards, Frederik Kriewitz

2 1

segfault error in bird6 after upgrading to 1.4.2
by Paul Emmons 18 Jun '14

18 Jun '14

Just upgraded to 1.4.2 from 1.3.9 and having issue with a ipv6 session that has 10k routes. Bird concierge as a route viewer ( no routing tables exchanged) Here is the error: Jun 17 17:21:30 collector kernel: [ 7107.186257] show_signal_msg: 3 callbacks suppressed Jun 17 17:21:30 collector kernel: [ 7107.186264] bird6[2295]: segfault at ffff800092721fa9 ip 00007fc03639dc2d sp 00007fff6d8ddfb0 error 5 in bird6[7fc036378000+70000] Jun 17 17:21:31 collector bird6: Started Jun 17 17:22:12 collector kernel: [ 7149.335707] bird6[15384]: segfault at ffff8000a4f155d9 ip 00007f0bdbc37c2d sp 00007fff5b0ea980 error 5 in bird6[7f0bdbc12000+70000] Jun 17 17:22:13 collector bird6: Started Jun 17 17:23:33 collector kernel: [ 7229.888236] bird6[15638]: segfault at ffff8000d3104cc9 ip 00007fbf1e824c2d sp 00007fff2cefb290 error 5 in bird6[7fbf1e7ff000+70000] Jun 17 17:23:33 collector bird6: Started Jun 17 17:24:00 collector kernel: [ 7256.689590] bird6[15710]: segfault at ffff8000c005edf9 ip 00007f69599a1c2d sp 00007fff3ffa1160 error 5 in bird6[7f695997c000+70000] Jun 17 17:24:00 collector bird6: Started Jun 17 17:25:43 collector kernel: [ 7360.327092] bird6[15741]: segfault at ffff8000a909a329 ip 00007f4195995c2d sp 00007fff56f65c30 error 5 in bird6[7f4195970000+70000] Jun 17 17:25:44 collector bird6: Started Jun 17 17:26:29 collector bird6: Reconfiguring Jun 17 17:26:29 collector bird6: Removing protocol Hurricane Stays up after I comment out users: #protocol bgp Hurricane { # description "as6939"; # local as 65000; # neighbor 2001:504:3b::14 as 6939; # route limit 35000; #} I've tried it with and without the route limit in it. There are <10k routes being sent. If I try to send that many routes (actually a full table) from another peer I do get the same error. It seems to be an issue with going over a certain number of ipv6 routes Thanks - Paul

2 1

check_mk bird plugin
by Frederik Kriewitz 18 Jun '14

18 Jun '14

Hello, our bird plugin for the check_mk monitoring system just made it to version 1.0. It's available here: https://github.com/freddy36/check_mk_extensions/tree/master/bird I tried to add it to the Related Software page of the wiki, but it seems like I can't do that. Best Regards, Freddy

2 1

Default route not picked up by OSPF?
by Grzegorz Nosek 16 Jun '14

16 Jun '14

Hi, I'm using bird to autoconfigure networking between a bunch of VMs running on a bunch of hosts. It basically looks like: - OSPF on every host (connecting host<->VMs) - OpenVPN in tap mode across all hosts - BGP running over that VPN The net result is that I have all VM private addresses propagated everywhere. It generally works fine except during host startup, where I regularly end up with VMs lacking the default route (which should have been supplied via OSPF). `birdc reload my_ospf` from the host makes the routes appear on the VMs, so I believe bird is configured correctly (i.e. it's only a transient issue, the routes aren't filtered out). The hosts have static default routes. OS is Ubuntu 12.04, using bird from the PPA (1.4.x). Is this a known issue (e.g. with daemon start order or whatever)? Any hints? I wish I had more diagnostics but like I said, it happens early in the boot process. Best regards, Grzegorz Nosek

1 0

Modification to BIRD to support better anycast model
by Jay Grizzard 10 Jun '14

10 Jun '14

Hiiiii bird folks. So, I've been looking at setting up a relatively simple anycast config, your typical 'multiple servers broadcast the same route via ospf and network proximity determines which you route to' type setup. BIRD seems like it should be able to do this fairly well, and seems to be one of the better/cleanest designs available for this sort of thing. I do have one issue, though, which seems to be common to all of the software-based routing solutions out there: Ongoing service health isn't really required for continued routing. The typical setup is to do something like monitor the service health with a local tool, and withdraw the route if that tool discovers lack of health. The problem with this is that if something goes wrong with that check -- it hangs, cron dies, or whatever -- the route will keep getting broadcast and traffic will keep getting directed to a system no longer able to handle it. So, I'd like to come up with a basic solution to this, where something active is required to keep anycast routes broadcasting, so if the app, monitoring, cron, etc die, the route will stop getting broadcast, without the monitoring having to 'manually' remove the route. My basic idea right now, in the spirit of "minimum workable footprint," is to do something like having a route filter tied to a file on disk, where the timestamp on that file is used to determine if the route should be exported... if the file ages beyond a certain point, the route gets filtered. So, something like this (syntax totally not thought through yet): filter ospf_out { if (net == 1.2.3.0/24) then { if fileage("/some/path") > 5 then reject; accept; } } ... and then have a health check that checks the app health, and touches /some/path. If the app goes down, or something goes wrong with the monitoring itself (or it just doesn't run at all), that file doesn't get touched, route gets filtered. This seems like the minimum viable interface for making this happen, especially as far as needing to make changes to BIRD is concerned -- it's pretty trivial to read the timestamp on a file on a repeating basis. My questions for the list, then: - Am I correct that BIRD doesn't currently have any functionality of this type available? I've gone through the manual a couple of times and don't see anything obvious, but I could have missed something. - Assuming it can't, where would be the right place to implement this? Is my guess about using an export filter a reasonable one? Anyone with code familiarity care to comment on any gotchas I might run into while doing this? - Anyone know of any non-BIRD tools that can do what I want natively? I looked at the obvious stuff (e.g. quagga) and didn't see anything, but would love to be corrected on this. Thanks! -jay

3 2

preferred route statistics
by Frederik Kriewitz 10 Jun '14

10 Jun '14

Hello, I noticed that the preferred route counter seems to be incremented for each time the route is exported by another protocol. E.g. I've a static protocol with 25 routes which are exported by two other protocols, this results in the following statistics: 25 imported, 0 exported, 50 preferred I'm wondering if that's the intended behavior? So far I expected that preferred means number of routes which are considered "best" in the corresponding routing table (so preferred would be always <= imported). Best Regards, Freddy

2 1

[PATCH] Maximum include depth in Bird config
by Artyom Gavrichenkov 10 Jun '14

10 Jun '14

Hi all, The page at http://bird.network.cz/?get_doc&f=bird-3.html , section 3.2 (as well as doc/bird.sgml), states that "the maximal depth [of inclusion of a new file in a Bird config] is 5". However, effective depth since July, 18, 2012, is 8 (9eceab33:conf/cf-lex.l:82). Maybe the user's guide needs an update, or depth 8 is unsupported? Just in case, patch goes below. Signed-off-by: Artyom Gavrichenkov <ximaera(a)qrator.net> --- doc/bird.sgml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/bird.sgml b/doc/bird.sgml index fa4c777..beacd4b 100644 --- a/doc/bird.sgml +++ b/doc/bird.sgml @@ -318,7 +318,7 @@ protocol rip { <p><descrip> <tag>include "<m/filename/"</tag> This statement causes inclusion of a new file. <m/Filename/ could also - be a wildcard. The maximal depth is 5. Note that this statement could be + be a wildcard. The maximal depth is 8. Note that this statement could be used anywhere in the config file, not just as a top-level option. <tag><label id="dsc-log">log "<m/filename/"|syslog [name <m/name/]|stderr all|{ <m/list of classes/ }</tag> -- 1.8.1.2

2 2