[Bug]: Investigate Kind IPv6 problem using HA proxy

[see-quick]

Bug Description

We are using a Kind cluster with 3 control planes. This means that for inter-communication between them, there is a need to provision a HA proxy (that's how kind is doing it). But it seems that our setup-kind.sh is falling due to

+ docker exec kind-cluster-control-plane /bin/sh -c 'echo "fd01:2345:6789::1    myregistry.local" >> /etc/hosts'
+ for node in $(kind get nodes --name kind-cluster)
+ echo 'Executing command in node:kind-cluster-external-load-balancer'
Executing a command in node:kind-cluster-external-load-balancer
+ docker exec kind-cluster-external-load-balancer /bin/sh -c 'echo "fd01:2345:6789::1    myregistry.local" >> /etc/hosts'
OCI runtime exec failed: exec failed: unable to start container process: exec: "/bin/sh": stat /bin/sh: no such file or directory: unknown

inside HA there is no sh but for IPv4 this is passing.
There are a few possible approaches how to solve this issue:

1. Use only 1 control plane (remove the need of HA proxy to be deployed)
2. Investigate if we need to add a DNS record to a /etc/hosts file. If not we could simply filter it.

Steps to reproduce

1. Provision a machine (Linux-based)
2. install docker
3. export IP_FAMILY=ipv6
4. install ./setup-kind.sh

Expected behavior

No response

Strimzi version

0.40.0 but I think this also applies on 0.39.0 and <=

Kubernetes version

kind version 0.22.0 with kindestnodes v1.29.2

Installation method

No response

Infrastructure

Kind

Configuration files and logs

No response

Additional context

No response