As this windows, have you checked to see if the network adaptor switched to being a “Public Network” or “Unknown Network” - that happens sometimes on windows and would need user intervention to set it back to “Private Network”
Also if you are still having the problem, see if the port forward to 192.168.0.156 for wan port 32400 is still in place and forwarding to local port 32400
You could test the port forward (whilst plex media server is running) by trying http://1.159.xxx.xxx:32400/web in browser from outside the local network
What firewall do you have? Any additional security product?
Thanks - I turned on Windows Firewall again (turned it off before) and i can see the Plex server while at work. As soon as i turned it back on it become visible? Network settings were set to Private.
Also, one of my kids is watching a movie from another room and after 40mins it seems fine.
However, it still is showing that it is not accessible remotely - but it all is working?
If I read this right and the URL works from my device connects on cellular, it means the red exclamation is a bug then? I was working well when using my Synology NAS as a server last week, but I recently fired up a older PC (Win 7) to handle the TV turner performance and I’ve had this error from the moment I installed. The Commercial Skip app is also crashing, but that’s for another message thread. For the record, I have had access without issue once I setup proper port forwarding to the correct device on my firewall. I just have a red exclamation point.
Unlike @jmosky my logs are showing successful connectivity tests, followed pretty quickly by unsuccessful tests which correspond to when the remote access indicator changes from green to red.
I can also access my plex server remotely, the remote access indicator just shows an incorrect status
Yes we are losing connectivity test requests and that is when you see the connectivity="0" events - I have referred this to our plex.tv team to look into possible loss of requests from the back end system. If there is any router / firewall log that confirm these requests are not coming through when others are then that would help.
Need to establish if there is evidence of any overload at the back end test systems - the full log would show me if there are any delays in receiving the events and if any rate limiting on tests is reached
Due to the way my firewall is configured, I’m only able to log passed packets and not rejected or blocked packets. However, comparing the timestamps of the notifyConnectivity message with the firewall time stamps it appears that both the connectivity="0" and connectivity="1" appear to be getting passed through to PMS.
I also noticed that the sometimes the Remote Access toggle swtich would changed from enabled to disabled and I would need to reactivate it again. It dropped out and changed to relay while I was doing the test. Hopefully that shows up in the logs.
Thanks for the logs and also the firewall log which is very useful - but the number of requests in the firewall log do not match what is in the server log.
The server log shows 240 WAN requests coming in during the period but the firewall log only has 67
I need to match requests
In the last server log in the zip which covers period:
Apr 19, 2020 21:29:06 to Apr 19, 2020 21:41:00
The firewall log covers:
Apr 19 21:29:33 to Apr 19 21:40:55
And I would like to establish somehow if you have a duplicate IP on the network. Could you explore that - the IP of interest being 192.168.2.230
There should be network tools available to detect that.
I can see requests in the firewall log but not in the server log (and vice versa - but that is different)
And one more thing - please leave some time between disabling/re-enabling remote access and then repeating it again, so that the diagnostics can be clear - there are time delays and it would be good if you leave remote access enabled for a number of minutes before disabling/re-enabling or clicking on reconnect.
You can see from here that the only good period I have for diagnostics is between 21:32:58 and 21:35:26 where we have a few minutes to look at.
Have checked the ARP table on my managed switch and firewall. Have also checked networking configuration on my docker host. There are no duplicate entries.
I belive the firewall is only logging the initial connection not each packet. I’ve run a packet capture on all WAN traffic with prt 32400 and will send that to you togther with the PMS logs.
I was not manually disabling then renabling remote access. I enabled it once, and then PMS was changing the status on it’s own. The only time I enabled it manually was at the start and then when it disabled on its own.
Playing around with it this morning, the remote access status stays red when I enable it from a locally based browser (i.e. a PC in my LAN(. If I then use a remote web browser to access PMS (in this instance my mobile phone), the remote status will switch to green. Sometimes I need to use the manual retry button next to the port setting from the remote browser. Provided I’m accessing PMS through my remote browser the status stays green. If I then access PMS through my local browser the status will switch to red again. I also noticed if I used the Plex app on my Android mobile, the status in the remote browser would switch to indirect, and remote access was showing as disabled rather than just not available. This also began to happen with my remote browser as well. The change is status also happened when I used the Plex appp on my local PC.
So the takeaway point for me is that the remote access status seems to change depending on whether I’m access my PMS locally or remotely.
I’ll forward the detailed logs to you via PM shortly.
At the moment from what I have seen, this is the only feasible explanation. some of the requests getting routed to a different host from the port forward. I do not expect there to be a Plex Media Server bug where we would lose tcp requests coming in. Could you also send me your docker configuration file.
It should not do it iself - were you perhaps clicking on the “RETRY” button on the manually specify public port line in remote access advanced settings ? The “RETRY” simulates a disable/re-enable.
I’ve forwarded the information. My docker is running plex using ipvlan rather than bridged networking so it should be exposed directly to my LAN subnet rather than relying on a port forward accross the docker bridge. Docker is running on CentOS which is within a VM on a physical Hyper-V host.
In some instances I was clicking retry when it changed, in others it changed status or disabled by itself without my manual intervention. I’ve just sent you detailed logs via PM with the logs, firewall packet capture and list of actions that I took. These were through a web browser via plex.tv/web:
7:24 Started plex - Remote access status red
7:33 Attempted Retry buttom from a web brower on my LAN - Remote access status remained red
7:37 Connected from brower on my mobile via WAN interface to PMS. Went to remote access screen and status changed green (did not select retry or change any other configuration settings
7:43 On my LAN broswer changed from remote access screen and back to PMS home screen – remote access status changed to red.
7:45 Connected from mobile broswer to WAN - Remote access stayed red
7:46 Attempted Retry buttom from mobile web brower on WAN – Status went green
7:50 On my Lan broswer changed from remote access screen and back to home – remote access status changed to red.
8:00 Used Plex App on mobile, noticed that connection type on my mobile brower connecting from WAN changed to indirect
8:02 Retried connection from mobile browser on WAN, no connection available
8:03 Checked status from browser on LAN and remote access has been disabled (I did not disable it manually).
8:08 Enabled Remote Access again from LAN browser – Remote access status remained red
8:14 Re-entered plex.tv/web in my mobile brower and it reconnected from WAN. Connection status switched from Indirect to Remote, but still shows remote access as red.
8:15 Navigated on mobile browser on WAN to PMS home then back to remote access configuration screen and status changed to green.
8:20 Navigated on mobile browser on WAN to PMS home then back to remote access configuration screen and status stayed green.
8:23 Started Plex Windows App on same local PC - No immediate change in remote access status, but then changed to red shortly after.
8:29 From browser on LAN hit retry for remote access - Status changed to green
8:43 From mobile browser on WAN accessed PMS - Status changed to red
8:49 From browser on LAN hit retry for remote access - Status remained red
8:53 From mobile broawed on WAN hit retry for remote access – Remote access status red with PMS connection status now showing as indirect for the mobile browser
8:57 Re-entered plex.tv/web in my mobile brower and it reconnected from WAN. Connection status switched from Indirect to Remote, but still shows remote access as red.
9:08 From mobile broawed on WAN hit retry for remote access – Remote access status red
The actual network packet capture pcap file with the server logs was very useful. We are not losing connectivity test requests but some are aborting during TLS 1.2 dialogue with the server. I have referred this to our plex.tv team. Thank you for all then diagnostics and I will let you know if we need any more tests
