Guys,
Running Plex on CentOS 7 15.03 VM behing pfSense, with those ports open:

Manual port is enable = 32400 on the Remote Access setting. When I click retry to try to enable the access, I get:
for a second and changes to:
Any ideas?
Thanks!
Yes, upnp is an option, but not a safe one. With upnp enabled, I won’t have control of what is been opened from within the LAN.
As per Plex documentation, all ports needed are open and NATed to the correct IP. In fact it connects, make itself available, and than loses that connection.
In fact, while I’n typing this the server is again available outside without any changes.
What of the logs would indicate when the server tried to open remote access? I just saved them and will look around 22:29 EST time stamp to see if I can figure out.
At the bottom the page I posted, you have the ACL list.
Denying by default to all and then allowing only your PMS server(s) is how I’ve done it.
There is no question about other hosts on my LAN. With regards to the computers (PMS servers) themselves, both are Linux and I have full control over what’s on them and how they behave.
In your logs, you’ll see the attempts to map UPNP and NAT-PMP
The hour is late here. I will look again over the weekend. I have had the pfSense for just over 1 week now and am still learning it’s intricacies but am very impressed with both how easy it is to use and what it can do. I’ve already pushed the modem into RFC bridge mode and transferred all PPPoE into it thereby eliminating the double NAT issue. The rest of my LAN is also flat. It is VLAN’d appropriately to isolate Wifi from the core systems.
Here’s how the rest of my UPnP service is configured as reference.
You can also disable certain facets of the UPnP service such as automatic outbound NAT and have full control.
Thank you for your help. Good point about deny all and only allow Plex to do upnp.
For reference: Plex is running on FreeBSD 11 -> bhyve -> CentOS 7 15.01 and pf is also a VM under the same server.