Same issue as @jester_tenet. Was having the issue with the 172.17.0.1 address being picked up as my public IP. Restarted PMS and now I’m getting “Private Unknown IP; Public Unknown IP”.
Using a manually specified port with port forwarding properly configured and a static IP. Setup has been this way for years and haven’t had issues until a day or two ago when the incorrect IP began showing up.
Edit: Further testing and it appears that clients outside the local network are now incapable of connecting to my server (I’m assuming due to PMS not being able to pick up local/public IP). Local clients are fine. Whatever was done to “fix” this issue has clearly made things worse. Same exact behavior is being reported on Reddit and in another thread on this forum.
Problems can only be investigated through diagnostics / logs. If the remote access settings page is showing unknown IP addresses then that would explain why remote apps cannot connect.
check that the network interfaces are available.
if there is an IP change then the remote mobile apps may need to closed down and re-launched to pick any change
If the remote access settings page is showing unknown IP addresses then that would explain why remote apps cannot connect.
Obviously, yes. I said as much.
check that the network interfaces are available.
if there is an IP change then the remote mobile apps may need to closed down and re-launched to pick any change
I’m really struggling with how to respond to this. Yes - my network interface is up and accessible. No, my IP has not changed. Yes, I’ve restarted my remote clients.
You did see the part where you have multiple users in multiple threads reporting the same problem over the last 24-48 hours, yes?
Debug logging is enabled on my server but I’m leery of providing logs as they tend to contain quite a bit of info about my network and setup. I’ll look through and see if I feel comfortable enough to provide them.
Having lots of people say there is a problem does not help getting problems fixed. Problems can only be fixed by identifying them first. They can only be identified through detailed investigation of diagnostic data.
The 172.17.0.1 issue was only fixed 4 hours ago. And a change of IP from 172.17.0.1 to the real IP is still an IP change - not a real one but an IP change as far as client apps are concerned
Also from past investigations - a lot of “Me Too” cases end up to be different issues when investigated
As @Insanewords, static ip and no issues for ages. Around the 21st, started to get “indirect connections” being thrown up, now I have an iPhone client which is registering the server as being off line, whilst other clients (physically next to it) are connecting fine (local network). Same device can connect to the public IP with the web interface.
Unable to add the server manually, error code 401.
Can’t give you logs - no logs for something which doesn’t connect, plus too much info in the logs (please provide script to anonymise?)
Is there any way to tell if clients (non-web) are going indirectly? It doesn’t seem obvious (which may mean they aren’t?)
@sa2000 Thanks for the update and your continued attention to the matter. If I need to manually anonymise my logs I can go that route to help pinpoint an issue but I’m hoping I won’t have to do that.
The second issue on plex.tv has now been fixed as well.
It also related to the public IP detected for the server. This issue led to symptoms of enabling remote access getting stuck or the public IP shows as unknown on the setting page
Having this same issue with a VPS. Should I create my own thread? It’s maddening because now I keep getting locked out of my server settings (even when directly connecting to the server!). As soon as I “Claim” my server, it disappears from the server list, and even accessing it directly/from SSH tunnel localhost, it won’t even show up until clearing PlexOnline entrees from Preferences.xml.
Attached logs here.
Am running 1.13.8.5395, just updated to try fixing this issue.
Running default port, which is wide open according to any of the online port checking tools, also opened all the recommended ports off the support pages.
Can’t see any logs. I have been investigating a problem which appears to affect servers that have a WAN IP address as the local IP address and access to the server gets lost when remote access is disabled. I need logs to confirm if it is the same problem
Assuming you have not disabled debug logging, please locate the logs folder and copy it out and zip the copy and attach. Please do that after restarting the server and waiting 5 minutes.
(If debug logging is disabled you will need to edit the Preferences.xml file to add it - lets check the logs first)
Please see my post above that mentions a new issue has just been identified affecting servers with a WAN IP address. Please follow the advice in my post to get the logs so i can confirm if it is the same issue
There may be several issues here - one being the one I mentioned where one can get locked out if remote access is disabled when the local IP is a WAN IP
The logs shows a problem where the network interface is not being detected.
Talked with my VPS support. They were able to add a veth interface, but they can’t swap it out for the venet one. So I have a veth interface that only has a ipv6 IP (and it doesn’t appear public).
Is there a way to just get Plex to use the venet interface? This worked before, and it appears it’s decided to stop. I know it’s publicly available, since I can access the server through the public IP at this port.
VPS support just asked me to switch to the much more expensive KVM server to get a traditional eth interface.
