The customer wanted to enable the BFD protocol in the OVN pods but after setting bfd=enable the configuration vanishes from the pod.

OCP 4.18.21

ovn-nbctl --version
ovn-nbctl 24.09.3
Open vSwitch Library 3.4.3
DB Schema 7.6.0

Enabled the BFD: (was necessary to issue the command two times to enable it)

ovn-nbctl -v set Logical_Router_Port rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com options:bfd=enabled

Checked the config (it will vanish in a few minutes)

[root@master-0 ~]# ovn-nbctl list Logical_Router_Port rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com
_uuid : 907598d9-ec50-4946-8d27-66a60f160d9f
dhcp_relay : []
enabled : true
external_ids :

{gateway-physical-ip=yes}
gateway_chassis : []
ha_chassis_group : []
ipv6_prefix : []
ipv6_ra_configs : {}
mac : "fa:16:3e:e3:39:db"
name : rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com
networks : ["10.0.89.230/21"]
options : {bfd=enabled}
peer : "10.0.94.79"
status : {}

Also included a peer IP address but it it does no work (using FRR installed through dnf)

[quickcluster@upi-0 cgroup]$ sudo vtysh

Hello, this is FRRouting (version 8.5.3).
Copyright 1996-2005 Kunihiro Ishiguro, et al.

upi-0.clu04238914v2.lab.upshift.rdu2.redhat.com# show bfd peer
BFD Peers:
peer 10.0.89.230 vrf default (peer is the IP from the ovn pod)
ID: 3062223890
Remote ID: 0
Active mode
Status: down
Downtime: 5 day(s), 22 hour(s), 59 minute(s), 4 second(s)
Diagnostics: ok
Remote diagnostics: ok
Peer Type: configured
RTT min/avg/max: 0/0/0 usec
Local timers:
Detect-multiplier: 3
Receive interval: 300ms
Transmission interval: 300ms
Echo receive interval: 50ms
Echo transmission interval: disabled
Remote timers:
Detect-multiplier: 3
Receive interval: 1000ms
Transmission interval: 1000ms
Echo receive interval: disabled

peer 10.0.94.79 vrf default (peer is the own IP from the node)
ID: 3482460252
Remote ID: 3482460252
Active mode
Status: up
Uptime: 3 day(s), 3 hour(s), 41 minute(s), 5 second(s)
Diagnostics: ok
Remote diagnostics: ok
Peer Type: configured
RTT min/avg/max: 0/0/0 usec
Local timers:
Detect-multiplier: 3
Receive interval: 300ms
Transmission interval: 300ms
Echo receive interval: 50ms
Echo transmission interval: disabled
Remote timers:
Detect-multiplier: 3
Receive interval: 300ms
Transmission interval: 300ms
Echo receive interval: 50ms

Ping work from both sides:

From pod to node:

ping -c 4 10.0.94.79
PING 10.0.94.79 (10.0.94.79) 56(84) bytes of data.
64 bytes from 10.0.94.79: icmp_seq=1 ttl=64 time=2.16 ms
64 bytes from 10.0.94.79: icmp_seq=2 ttl=64 time=4.85 ms
64 bytes from 10.0.94.79: icmp_seq=3 ttl=64 time=0.241 ms
64 bytes from 10.0.94.79: icmp_seq=4 ttl=64 time=0.324 ms

— 10.0.94.79 ping statistics —
4 packets transmitted, 4 received, 0% packet loss, time 3034ms
rtt min/avg/max/mdev = 0.241/1.894/4.849/1.870 ms



From node to pod:

ping -c 4 10.0.89.230
upi-0.clu04238914v2.lab.upshift.rdu2.redhat.com# exit
[quickcluster@upi-0 cgroup]$ ping -c 4 10.0.89.230
PING 10.0.89.230 (10.0.89.230) 56(84) bytes of data.
64 bytes from 10.0.89.230: icmp_seq=1 ttl=64 time=6.04 ms
64 bytes from 10.0.89.230: icmp_seq=2 ttl=64 time=5.06 ms
64 bytes from 10.0.89.230: icmp_seq=3 ttl=64 time=0.534 ms
64 bytes from 10.0.89.230: icmp_seq=4 ttl=64 time=0.312 ms

— 10.0.89.230 ping statistics —
4 packets transmitted, 4 received, 0% packet loss, time 3021ms
rtt min/avg/max/mdev = 0.312/2.986/6.043/2.587 ms




After configuring the BFD it vanishes:

ovn-nbctl list Logical_Router_Port rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com
_uuid : 907598d9-ec50-4946-8d27-66a60f160d9f
dhcp_relay : []
enabled : true
external_ids : {gateway-physical-ip=yes}

gateway_chassis : []
ha_chassis_group : []
ipv6_prefix : []
ipv6_ra_configs : {}
mac : "fa:16:3e:e3:39:db"
name : rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com
networks : ["10.0.89.230/21"]
options : {}
peer : "10.0.94.79"
status : {}
[root@master-0 ~]#

And no routes with bfd marked:

IPv4 Routes
Route Table <main>:
169.254.0.0/17 169.254.0.4 dst-ip rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com
10.128.0.0/14 100.64.0.1 dst-ip
0.0.0.0/0 10.0.95.254 dst-ip rtoe-GR_master-0.clu04238914v2.lab.upshift.rdu2.redhat.com
[root@master-0 ~]# ovn-nbctl -v

The customer confirms that one cluster works (named ocp7) and the other one does not (ocp22).

Must gathers in the ticket 04262686 will be shared in Google Drive link.