Is this a bug report or a feature request?:
bug
What happened:
after changing config map and the restart we are still using the old addresses with the new deployments and the old ones.
What you expected to happen:
be able to get new addresses
How to reproduce it (as minimally and precisely as possible):

Anything else we need to know?:
To connect https://linkedin.com/in/rifaterdemsahin/
Environment:
uname -a):Also pod deletion does not change the outcome

also asked the related question here in reddit
https://www.reddit.com/r/kubernetes/comments/a0kevu/cant_access_external_ip_other_than_the_kubernetes/
MetalLB rejects new configurations if the new configuration would break existing services. Your new configuration does not allow existing services to continue existing, so MetalLB ignored it. There are log entries in the controller and speaker pods about this.
To force MetalLB to accept an unsafe configuration, delete all the controller and speaker pods. When they restart, they'll accept the new configuration and change all your services. kubectl delete po -n metallb-system --all
Most helpful comment
MetalLB rejects new configurations if the new configuration would break existing services. Your new configuration does not allow existing services to continue existing, so MetalLB ignored it. There are log entries in the controller and speaker pods about this.
To force MetalLB to accept an unsafe configuration, delete all the controller and speaker pods. When they restart, they'll accept the new configuration and change all your services.
kubectl delete po -n metallb-system --all