Sort of, it goes to the cloud for initial routing then the client/server talk directly.
Let me repeat what have already said: Plex expects a RFC 1918 LAN network. having a non-complaint LAN will cause plex to act weird. As a security measure Plex treats LAN and WAN traffic differently. Plex uses the RFC standards for identifying which LAN is vs WAN. When you dont follow network standards expect stuff to break. IE This is why plex is broken.
EDIT:
Yes, Plex enhanced their security which is what I have been saying since the start.
Ok. But, I still have the confusion as PLEX on my older server was happily doing its thing. Is this something new to PLEX? The old server was maybe not quite up to date, but was not running some archaic build.
You’ve given me an answer, so many thanks for that.
I’ll have to see what I can do with that as I am not really wanting to re-allocate my entire network. I could connect the server directly to the router which is serving up a 192.x.x.x private address, but there are pros/cons with that. I’ll give it a shot.
The internet is a funny place. I’m just asking questions and trying to fill the gaps in my knowledge against the input you are providing me. To me, this is not an argument, but a discussion. But as they say, it’s hard to read facial expressions on the internet
Could you perhaps describe your network topology/configuration in greater detail? I feel like there’s some we’re all missing. If you already have a router that’s terminating your public address(es) and NAT’ing them, that might mean you have your PMS in a DMZ.
A greater understanding of what you have, and how you have it configured, and what you’re attempting to accomplish would like help.
If you own the block, you can obviously use it however you’d like. But there are some applications/services which expect to be hosted primarily locally. PMS is one of those. I mentioned some exceptions above, but those need to be handled in fairly specific ways.
It’s connected to a larger switch that serves up all IP’s on the private block.
All ports are blocked on the modem except for a few specific purposes. Forwarding rules setup to the switch which forwards to the server.
A few routes set up where I need to route traffic to the modem.
When I get a chance, I am going to patch the server direct to the modem so it gets a 192.x.x.x, and is no longer double NAT’d and see what happens (probably on the weekend). I’m hoping that once activated, I can put it back on the private address and it will retain its authorized state.
As for my block. I use it, basically because I own it (sentimentality). Reluctance on switching back to 192.x.x.x is also basically because I don’t want to spend the time to setup all my static IP’s, reboot the plethora of devices, etc, particularly, before I see if there is another way.
Okay, to close the loop on this, and for anyone in the future who is having a similar problem.
Putting the new server on the modem network (private IP, single NAT) allowed it to authorize with no issues at all.
After authorization, putting the server back on the public/double NAT location it continued to be authorized and work.
So, this means that either (or both) the double NAT or public IP was the problem. I did not tinker with my network configuration to rule out which was the ultimate cause.
