- Bird-users - Mailing Lists

Re: BFD not recovering after link recovers
by 0.bgp＠elloe.vision 28 Jan '21

28 Jan '21

We added in backtrace()/backtrace_symbols() output from various error logging sources to understand the call stack. This shows that the cause of the final error is returned via bfd_open_tx_sk() from: sysdep/unix/io.c::sk_open(): sockaddr_fill(&sa, s->af, bind_addr, s->iface, bind_port); if (bind(fd, &sa.sa, SA_LEN(sa)) < 0) ERR2("bind"); } However, we're not clear on why, when the link comes back up, bird is unable to return to Up status. In theory once systemd-networkd reassigns IP addresses to the interface and adds back routes bird should be able to recover. We do not understand the code enough at present to determine what should be happening, or where to look to figure out a solution. Pointers to where we should be looking and what we should expect are welcome. bfd1: Socket error: bind: Cannot assign requested address Stack trace 13 frames /usr/sbin/bird(print_trace_io+0x34) [0x564255163b64] /usr/sbin/bird(bfd_open_tx_sk+0x12c) [0x5642551243ac] /usr/sbin/bird(+0x5e9c3) [0x5642551219c3] /usr/sbin/bird(+0x5ec98) [0x564255121c98] /usr/sbin/bird(bfd_request_session+0x80) [0x564255121ee0] /usr/sbin/bird(+0x6d77e) [0x56425513077e] /usr/sbin/bird(+0x71575) [0x564255134575] /usr/sbin/bird(+0x4e75c) [0x56425511175c] /usr/sbin/bird(ev_run_list+0xa1) [0x564255102441] /usr/sbin/bird(io_loop+0x5c) [0x564255166f1c] /usr/sbin/bird(main+0x786) [0x5642550e08c6] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf3) [0x7fb4e2c310b3] /usr/sbin/bird(_start+0x2e) [0x5642550e0a8e] > On Wed, Jan 27, 2021 at 03:23:25PM +0000, 0.bgp(a)elloe.vision wrote: >> We are setting up a container based proof of concept with IPv6 only >> using ECMP, ANYCAST, BGP with BFP and Bird2 and I've hit a problem >> where BFD doesn't recover after a local link goes down and comes back >> and are seeking some advice as whether this is expected behaviour or >> a bug. > Hi > > Based on quick evaluation, seems to me that it is a combination of > race condition in BIRD and systemd-networkd behavior. There is a > condition in BIRD that when a BFD session is added during the time IP > address on that link is added/removed, sometimes it fails to add the > socket. > > Seems like this issue is exaggerated by systemd-networkd, as normally > it happens just during admin-up/down events, but from the logs it > seems that as a reaction on link-up/down, systemd-networkd > adds/removes the IP address.

1 0

show memory question
by Alarig Le Lay 25 Jan '21

25 Jan '21

Hi, On one of my routers, I needed to free some RAM temporally, so I shut all my BGP sessions. With `birdc show memory` I saw that it should have worked, however htop was still seeing 1.2G used by bird. 5311 root 20 0 1296M 1220M 1724 S 0.0 61.1 59h01:29 ├─ /usr/sbin/bird -s /run/bird.ctl -c /etc/bird.conf - 5314 root 20 0 1296M 1220M 1724 S 0.7 61.1 6h31:09 │ └─ /usr/sbin/bird -s /run/bird.ctl -c /etc/bird.con bird> show memory BIRD memory usage Routing tables: 49 MB Route attributes: 842 kB Protocols: 300 kB Total: 50 MB So, I added the `disabled` option to my sessions and restarted bird, and then birdc and htop were almost consistent. 25724 root 20 0 107M 32600 2224 S 0.7 1.6 0:00.62 ├─ /usr/sbin/bird -s /run/bird.ctl -c /etc/bird.conf - 25726 root 20 0 107M 32600 2224 S 0.7 1.6 0:00.16 │ └─ /usr/sbin/bird -s /run/bird.ctl -c /etc/bird.con bird> show memory BIRD memory usage Routing tables: 28 MB Route attributes: 46 kB Protocols: 304 kB Total: 29 MB When I re-enabled the BGP sessions, the RAM utilisation was a bit lower (the original process was two months old). 25724 root 20 0 1147M 1072M 2072 R 1.3 53.7 2:33.80 ├─ /usr/sbin/bird -s /run/bird.ctl -c /etc/bird.conf - 25726 root 20 0 1147M 1072M 2072 S 0.7 53.7 0:07.84 │ └─ /usr/sbin/bird -s /run/bird.ctl -c /etc/bird.con bird> show memory BIRD memory usage Routing tables: 399 MB Route attributes: 528 MB Protocols: 102 MB Total: 1029 MB Perhaps bird thinks that some RAM has been freed while it’s not actually the case? -- Alarig

2 2

OSPFv2 and unnumbered interfaces without peer ?
by Etienne Champetier 25 Jan '21

25 Jan '21

Hi All, Is it possible to use unnumbered interfaces without having a peer address setup ? Without a peer the state stays 'PtP (stub)' # ip link add name wg-test type wireguard # ip addr add dev wg-test local 1.2.3.4 # ip link set wg-test up # birdc show ospf interface Interface wg-test (1.2.3.4/32) Type: ptp Area: 0.0.0.0 (0) State: PtP (stub) ... # ip addr del 1.2.3.4/32 dev wg-test # ip addr add dev wg-test local 1.2.3.4 peer 5.6.7.8 # birdc show ospf interface Interface wg-test (peer 5.6.7.8) Type: ptp Area: 0.0.0.0 (0) State: PtP ... Jason (in CC) and I were wondering if this is a Bird limitation or if having a peer setup is really important semantically (and worth adding to all protocols in OpenWrt)? https://github.com/openwrt/openwrt/pull/3810 Thanks Etienne

2 2

[bug] network prefix applied to local instead of peer address
by Etienne Champetier 25 Jan '21

25 Jan '21

Hi All, When configuring a ptp interface, the local address is /32, and the prefix is supposed to be applied to the peer address I not using a non /32 peer but was curious when reading `man ip-address` # ip link add name wg-test1 type wireguard # ip link set wg-test1 up # ip addr add dev wg-test1 1.2.3.4 peer 5.6.7.8/24 # ip addr show wg-test1 11: wg-test1: <POINTOPOINT,NOARP,UP,LOWER_UP> mtu 1420 qdisc noqueue state UNKNOWN group default qlen 1000 link/none inet 1.2.3.4 peer 5.6.7.8/24 scope global wg-test1 valid_lft forever preferred_lft forever # birdc show ospf interface Interface wg-test1 (1.2.3.0/24) Type: ptp Area: 0.0.0.0 (0) State: PtP ... (we should see `Interface wg-test1 (peer 5.6.7.8/24)`) # strace -f -s 10000 ip addr show wg-test1 recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base={{len=84, type=RTM_NEWADDR, flags=NLM_F_MULTI|NLM_F_DUMP_FILTERED, seq=1611526183, pid=25832}, {ifa_family=AF_INET, ifa_prefixlen=24, ifa_flags=IFA_F_PERMANENT, ifa_scope=RT_SCOPE_UNIVERSE, ifa_index=if_nametoindex("wg-test1")}, [{{nla_len=8, nla_type=IFA_ADDRESS}, inet_addr("5.6.7.8")}, {{nla_len=8, nla_type=IFA_LOCAL}, inet_addr("1.2.3.4")}, {{nla_len=13, nla_type=IFA_LABEL}, "wg-test1"}, {{nla_len=8, nla_type=IFA_FLAGS}, IFA_F_PERMANENT}, {{nla_len=20, nla_type=IFA_CACHEINFO}, {ifa_prefered=4294967295, ifa_valid=4294967295, cstamp=2331331, tstamp=2331331}}]}, iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 84 https://github.com/torvalds/linux/blob/master/include/uapi/linux/if_addr.h#… * IFA_ADDRESS is prefix address, rather than local interface address. * It makes no difference for normally configured broadcast interfaces, * but for point-to-point IFA_ADDRESS is DESTINATION address, * local address is supplied in IFA_LOCAL attribute. Best Etienne

2 2

Subject: FreeBSD/OSPF: ospf1: Socket error on wg0: Network is unreachable
by adam＠adammccarthy.co.uk 25 Jan '21

25 Jan '21

Hello, I run OSPFv2 between multiple Linux hosts over wireguard tunnels between them in a full mesh. This works fine in Ubuntu Linux 20.04, but if I try to add a FreeBSD 12.2-RELEASE-p1 host, I receive the following error: bird[9767]: ospf1: Socket error on wg-lon1: Network is unreachable Could this be related to changes in FreeBSD 12, like FreeBSD 11 e.g., https://github.com/BIRD/bird/pull/1? Thanks for any advice you can give me. Best, Adam BIRD and Wireguard are built from FreeBSD ports. $ bird --version BIRD version 2.0.7 $ wg --version wireguard-tools v1.0.20200827 - https://git.zx2c4.com/wireguard-tools/ $ uname -a FreeBSD <hostname> 12.2-RELEASE-p1 FreeBSD 12.2-RELEASE-p1 GENERIC amd64 wg0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> metric 0 mtu 1420 options=80000<LINKSTATE> inet 10.1.5.5 --> 10.1.5.5 netmask 0xfffffff8 groups: tun nd6 options=101<PERFORMNUD,NO_DAD> Opened by PID 9459 $ netstat -rn [...] 10.1.5.1 wg-lon1 UHS wg-lon1 10.1.5.5 link#5 UH wg-lon1 [...] $ cat /etc/bird/bird.conf [...] protocol ospf { ipv4{ import all; export all; }; graceful restart on; area 0 { interface "wg0" { type nbma; authentication none; neighbors { 10.1.5.1 eligible; }; }; }; } [...]

2 2

MultiBird on L2 - A crazy idea for Fail Over y and Load Balancing
by Douglas Fischer 24 Jan '21

24 Jan '21

I was studying the concepts of multi-bird for large environments of IXPs. And, beyond the extra complexity that it brings to the environment, one of the weak points I saw was the fact that all the Bird instances are at the same box(vm, container, etc...). A friend mentioned that some tests were made with a LoadBalancer redirecting the post-nated connections to other boxes. But even in that scenario, that load balancer would be a single-point-of-failure/bottleneck. So I was remembering Cisco GLBP and Heart-Beat protocol. Those protocols inform different Mac-Addresses to the same IPv4/IPv6 Address, based on the source of the ARP/ND query. Making a load-balance/fail-over based on the glue between layer2 and layer3. P.S.: Several scenarios uses that concept. Corosync, Windows Cluster, Orale RAC, etc... Considering that concept, and joining it with multibird: Would be possible to create groups of sources and assigning different priorities to those groups on each instance of Bird. In this case, each Bird instance could run on a different box, or even on a different site. Further than that, on IXPs with a large number of participants, would be possible to define some affinity between that group of priority based for example on the facility where those participants are connected. I have a feeling that this would be especially useful for remote peering scenarios. Just a crazy idea to share with colleagues. Maybe from here, some good thing could rise. -- Douglas Fernando Fischer Engº de Controle e Automação

5 13

How unbind bfd from ipv4 socket
by Дружок 1 22 Jan '21

22 Jan '21

Hello, All! I have two bird2.0.7 process run in same Linux system: bird bird6 (copy of sbin/bird) It used two different config files. In each config describes protocol bfd But I need that process “bird” unbind own bfd from v6 address, and process “bird6” - from v4 address (conflict). How do it? -- *-- * *-- С Уважением*

3 2

Converting a route between tables via pipe proto?
by David Johnson 22 Jan '21

22 Jan '21

Hi, I'm trying to see if bird can accomplish the following and so far I can't seem to find a way to do it. Quick summary: I want to be able to modify a route (the actual IP/mask) between 2 interfaces. Setup is as follows: table A: routes learned as normal table B: I want to advertise a specific route into table B if there is another specific route learned on table A. Note that the route in table B is different than A but I want to link/convert them somehow. Why would I want to do this? This Linux setup is performing NETMAP to convert IPs when forwarded between 2 interfaces. I know the exact prefixes so I can enter them in the config but I want to dynamically advertise those into B but only if they are reachable via A. My first attempt was to use a proto pipe between the two tables and simply change the net variable: protocol pipe a_b_pipe { table b; peer table a; import filter { # a -> b if net ~ A.B.C.D/E then { net = F.G.H.I/J; accept; }; reject; }; export none; # b -> a } but that doesn't work: bird: /etc/bird/bird.conf:24:56 This static attribute is read-only. My second attempt was to setup filters/functions on the protos in table A to set some variable if the specific route is seen, then reference that variable via a static proto route in table B. This seems to be a no go as there are global constants (define xxx;) but attributes are local and function variables are also local scope. My third thinking was to somehow use bfd to cross the bridge as it were.. setup a static route with the bfd option in table B and have it communicate with a bfd in table A, but there doesn't seem to be any way to start/stop bfd on the table B side based on learning a specific route in table B. Other suggestions appreciated, or is this just not possible?

3 2

Duplicated PEER_INDEX_TABLE in rit mrt file
by Santiago Aggio 22 Jan '21

22 Jan '21

Hello, I am using bird-2.0.7 on Debian 10 to dump the RIBs tables. The tables generated for IPv4 are fine when parsed using mrt2bgpdump or bgpscanner. But when I parse the IPv6 table, mrt2bgpdump doesn't return anything and bgpscanner gives the following error: bgpscanner: master6.mrt: bad RIB dump, duplicated PEER_INDEX_TABLE, skipping rest of file. Can you help me? Thank you in advanced. Santiago.

3 4

KRT: Error sending route 192.168.x.x/32 to kernel: File exists
by Victor Sudakov 20 Jan '21

20 Jan '21

Dear Colleagues, My system is full of the following messages: bird[59366]: KRT: Error sending route 192.168.246.1/32 to kernel: File exists bird[59366]: KRT: Error sending route 2001:470:ecba:2::1/128 to kernel: File exists bird[57195]: KRT: Error sending route 2001:470:ecba:4::1/128 to kernel: File exists vas bird[57195]: KRT: Error sending route 192.168.170.0/24 to kernel: Address already in use (as you see, these are not only host routes, but network routes as well). They seem harmless but clog the log. Why is Bird trying to install those routes? Is there an extraneous export or import statement somewhere? Or should I just quelch those warnings and forget? Here is my config with comments removed (bird2 2.0.7_1, FreeBSD 12.2), and the kernel IPv4 route table below it. log syslog all; router id 192.168.246.1; protocol device { scan time 10; } protocol direct { disabled; ipv4; ipv6; } protocol kernel kernel4 { ipv4 { export all; }; } protocol kernel kernel6 { ipv6 { export all; }; } protocol static { ipv4; } protocol ospf v2 ospf4 { ipv4 { import all; export all;}; area 0 { interface "re*" { cost 10; }; interface "bridge*" { cost 10; }; }; } protocol ospf v3 ospf6 { ipv6 { import all; export all;}; area 0 { interface "re*" { cost 10; }; interface "bridge*" { cost 10; }; }; } $ netstat -rn -finet Routing tables Internet: Destination Gateway Flags Netif Expire default 192.168.170.1 UG1 re1 127.0.0.1 link#3 UH lo0 192.168.89.0/27 192.168.170.1 UG1 re1 192.168.170.0/24 link#2 U re1 192.168.170.5 link#2 UHS lo0 192.168.246.0/24 link#4 U bridge0 192.168.246.1 link#4 UHS lo0 -- Victor Sudakov, VAS4-RIPE, VAS47-RIPN 2:5005/49@fidonet http://vas.tomsk.ru/

3 5