Plex docker shows IP of non-active eth interface

[fiore00713]

I have two 10GbE NICs attached to my unRAID box (running 6.5.2)

One of which is being used as direct connection between my primary unRAID box and a second unRAID server for taking backups to. This is a direct connection between the two servers with static IPs set, which are on a completely different subnet from my production network

 

I noticed that sometime over the weekend my Plex server was reporting the wrong IP address in the Remote Access area of Server settings. Oddly enough it is showing the 10. address that is assigned on the 10GbE (eth10) port.

 

In Dockers it is showing the correct IP, that of the unRAID server so I'm a bit confused about the discrepancy.

Everything APPEARS to be working correctly, though Remote Access states my server is not accessible outside my network (but I'm able to connect to it from mobile hotspot as well as shared access users being able to reach it) I'm just trying to determine what is going on.

 

For testing, I have disabled the eth10 port within Network Settings in unRAID and restarted the docker, this has resulted in no change. I have rebooted the server multiple times with no change. The one thing I have not done is physically disconnected the interface, though I didn't feel this should be necessary if I'm setting the port administratively down.

 

Attached are unraid diagnostic logs

I can provide Plex logs but I'm unsure which would be necessary

 

unraid-diagnostics-20180625-1612.zip

[ken-ji]

Is the Plex container in Host mode? Because that would let the Plex server see all the network interfaces of the host.

Otherwise, it should only be seeing either the private/internal docker subnet or the dedicated IP you've assigned to it (if you've enabled docker custom networks).

 

[fiore00713]

20 minutes ago, ken-ji said:

Is the Plex container in Host mode? Because that would let the Plex server see all the network interfaces of the host.

Otherwise, it should only be seeing either the private/internal docker subnet or the dedicated IP you've assigned to it (if you've enabled docker custom networks).

 

 

It is in host mode but that's the mode it's been in since I setup the server over a year ago. The 10GbE was added about 3 months ago

UNLESS I'm just noticing this (which is possible but I had not seen the Remote Access throwing an error)