[BFD] New loxilb instance not retaining MASTER state after failover

[nik-netlox]

Describe the bug
When Master loxilb instance is restarted then state switch is working fine but as the instance comes back again then it reclaims MASTER state before the connection sync completes. New Master does not have connection entries so it sends a RESET to the client.

To Reproduce
Running cicd/k8s-calico-ipvs2-ha-ka-sync/
llb1

$ sudo docker run -u root --cap-add SYS_ADMIN   --restart unless-stopped --privileged -dit -v /dev/log:/dev/log --net=host --name loxilb ghcr.io/loxilb-io/loxilb:latest --cluster=192.168.80.253 --self=0 --ka=192.168.80.253:192.168.80.252

llb2

$ sudo docker run -u root --cap-add SYS_ADMIN   --restart unless-stopped --privileged -dit -v /dev/log:/dev/log --net=host --name loxilb ghcr.io/loxilb-io/loxilb:latest --cluster=192.168.80.252 --self=1 --ka=192.168.80.252:192.168.80.253

Expected behavior
New Master loxilb must retain it's state when the other loxilb should get BACKUP state when it comes alive.

Screenshots

Environment (please complete the following information):

• OS: Ubuntu 20.04
• Kernel Version: 5.15.5
• LoxiLB Version: latest
• Cloud Environment: On-prem
• K8s Version [N/A]

Additional context
Add any other context or topology about the problem here.

[nik-netlox]

Verified. It is working.