Description of problem:
It seems in some situations cephadm may not setup services such as grafana to bind correctly to a specific network interface. This can cause a deployment failure as it tries to bind to all interfaces and conflicts with the related haproxy config.
More details in this cephadm bz: https://bugzilla.redhat.com/show_bug.cgi?id=2233659
The goal here is to determine if tripleo can better handle this config until cephadm bug is resolved; perhaps different ports for haproxy and backend services.
This seems to impact ipv6 deployments and multi-segment overcloud storage networks; perhaps other scenarios.
The typical ipv4 single segment storage network is not impacted.
Version-Release number of selected component (if applicable):
OSP 17.1
How reproducible:
See related bz and description above.