Host DNS broken when running a DNS server in bridge mode

Hi all,

I’m in the process of introducing Consul Connect to my Nomad cluster, and ran into issues with CoreDNS lately.
CoreDNS is running as a system job for quite some time now without issues in host mode, but when I try to use bridge networking, DNS breaks on the host, and by extension on all containers running on that host.
CoreDNS itself is running fine. Other hosts can access CoreDNS via :53.

Anyone here who had encountered similar issues and could lend me a hand what the root cause might be?

Nomad 1.7.1
Consul 1.17.0
Ubuntu 23.04
Docker 24.0.5