Got this strange error while staring bird. Do I need anything else to intsall ?

BIRD version: 2.0.8
OS: Arch Linux

Systemd logs:

Starting BIRD routing daemon...
/etc/bird.conf:18:5 syntax error, unexpected IMPORT
bird: /etc/bird.conf:18:5 syntax error, unexpected IMPORT
bird.service: Control process exited, code=exited, status=1/FAILURE
bird.service: Failed with result 'exit-code'.
Failed to start BIRD routing daemon.

The tutorials works fine when running on vagrant.
After replicating it on openstack VMs.The traffic is not going from openstack ingress VM to cluster.

all check points are working as mentioned in the example,

`
sudo birdc show protocols
BIRD 1.6.8 ready.
name proto table state since info
bgp1 BGP master up 2021-08-21 Established
bgp2 BGP master up 2021-08-21 Established
bgp3 BGP master up 2021-08-21 Established
kernel1 Kernel master up 2021-08-20
device1 Device master up 2021-08-20

sudo birdc show protocols
BIRD 1.6.8 ready.
name proto table state since info
bgp1 BGP master up 2021-08-21 Established
bgp2 BGP master up 2021-08-21 Established
bgp3 BGP master up 2021-08-21 Established
kernel1 Kernel master up 2021-08-20
device1 Device master up 2021-08-20

sudo birdc show route protocol bgp2
BIRD 1.6.8 ready.
172.16.251.128/26 via 192.168.0.22 on ens3 [bgp2 2021-08-21] * (100) [i]
sudo birdc show route protocol bgp3
BIRD 1.6.8 ready.
172.16.216.0/26 via 192.168.0.24 on ens3 [bgp3 2021-08-21] * (100) [i]
sudo birdc show route protocol bgp1
BIRD 1.6.8 ready.
172.16.137.0/26 via 192.168.0.5 on ens3 [bgp1 2021-08-21] * (100) [i]

sudo calicoctl node status
Calico process is running.
IPv4 BGP status
+--------------+-------------------+-------+------------+-------------+
| PEER ADDRESS | PEER TYPE | STATE | SINCE | INFO |
+--------------+-------------------+-------+------------+-------------+
| 192.168.0.21 | global | up | 2021-08-21 | Established |
| 192.168.0.22 | node-to-node mesh | up | 2021-08-20 | Established |
| 192.168.0.24 | node-to-node mesh | up | 2021-08-20 | Established |
+--------------+-------------------+-------+------------+-------------+

IPv6 BGP status
No IPv6 peers found.

Only difference is route command output on ingress node
route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
default _gateway 0.0.0.0 UG 100 0 0 ens3
169.254.169.254 _gateway 255.255.255.255 UGH 100 0 0 ens3
172.16.137.0 192.168.0.5 255.255.255.192 UG 0 0 0 ens3
172.16.216.0 192.168.0.24 255.255.255.192 UG 0 0 0 ens3
172.16.251.128 192.168.0.22 255.255.255.192 UG 0 0 0 ens3
192.168.0.0 0.0.0.0 255.255.255.0 U 0 0 0 ens3

There is an extra interface enp0s3: in virtualbox set up for each VM whereas openstack VMs have only
ens3 interface.

Is there something more which needs to be configured.

I've been testing this for a while now and I notice there's always a defunct haproxy process with haproxy-ingress-controller as parent. Is this expected behaviour?

haproxy   268787       1  0 15:10 ?        00:00:00 /usr/local/bin/haproxy-ingress-controller --external --configmap=haproxy-ic/kubernetes-ingress --program=/usr/sbin/haproxy --ipv4-bind-address=0.0.0.0 --http-bind-port=80 &
haproxy   268868  268787  0 15:10 ?        00:00:00 [haproxy] <defunct>
haproxy   268870       1  0 15:10 ?        00:00:00 /usr/sbin/haproxy -f /tmp/haproxy-ingress/etc/haproxy.cfg
haproxy   268919  268870  1 15:10 ?        00:00:11 /usr/sbin/haproxy -sf 268871 -x sockpair@3 -f /tmp/haproxy-ingress/etc/haproxy.cfg

Could it be that it's because of config testing of some kind, using the haproxy binary, and haproxy-ingress-controller isn't reaping it?

As I've seen e.g. this in ps output as well

 /usr/sbin/haproxy -f /tmp/haproxy-ingress/etc/haproxy.cfg.1f5499b6-1541-4e86-95bd-4aa99d1104be -c

Which is a config check, but that one does disappear from the process list.

Hello
how to configure haproxy for multi master cluster with haproxy-ingress-controller on the same time, I did not deactivate the haproxy service as you indicated in the tutorial because I use it as HA for multi controleplane. the HA service is running but HA-ingress-controller is not.
for the Bird.conf configuration I assigned the 3 controlplanes.

controlplane1

protocol bgp {
local 192.168.50.21 as 65000;
neighbor 192.168.50.22 as 65000;
direct;
import filter {
if ( net ~ [ 172.16.0.0/16{26,26} ] ) then accept;
};
export none;
}

controlplane2

protocol bgp {
local 192.168.50.23 as 65000;
neighbor 192.168.50.22 as 65000;
direct;
import filter {
if ( net ~ [ 172.16.0.0/16{26,26} ] ) then accept;
};
export none;
}

controlplane3

protocol bgp {
local 192.168.50.21 as 65000;
neighbor 192.168.50.24 as 65000;
direct;
import filter {
if ( net ~ [ 172.16.0.0/16{26,26} ] ) then accept;
};
export none;
}

worker1

protocol bgp {
local 192.168.50.21 as 65000;
neighbor 192.168.50.25 as 65000;
direct;
import filter {
if ( net ~ [ 172.16.0.0/16{26,26} ] ) then accept;
};
export none;
}

worker2

protocol bgp {
local 192.168.50.21 as 65000;
neighbor 192.168.50.26 as 65000;
direct;
import filter {
if ( net ~ [ 172.16.0.0/16{26,26} ] ) then accept;
};
export none;
}