Issue with XFS Docker disk format and docker port binding

I just changed my docker disk format from BTRFS to XFS due to heaps of errors in my unraid log showing up. Quick google showed a rebuild of the disk was the easiest way to resolve (ignore) it.
I stopped the docker service, chose an XFS disk format instead and started it again.
I have a user made bridged docker network where all of my dockers have access to so they are able to send logs to my log collector even if they don't have an external IP.
Everything was all good up until I tried to start a PostgreSQL docker when there was already one started for another web service. Then I got the error

Quote

Bind for 0.0.0.0:5432 failed: port is already allocated

Intrigued why changing only the format of the disk caused this I jumped back to the old disk to confirm I wasn't going crazy and both of the PostgreSQL dockers started up perfectly fine with no port conflicts.

They are both being composed using the exact same docker stack, although each in a different stack.

I then did a docker inspect on the same docker, composed in both the BTRFS and XFS disk formats, and here is the main difference (minus different ID's etc).

The BTRFS 'Ports' section is empty, but the XFS section has the following.

            "Ports": {
                "5432/tcp": [
                    {
                        "HostIp": "0.0.0.0",
                        "HostPort": "5432"
                    },
                    {
                        "HostIp": "::",
                        "HostPort": "5432"
                    }
                ]
            },

Am I running into a bug? Or have I been using a bug with BTRFS and XFS is the correct way it should be?