I’m not sure about those times. I was not looking at the system then (that I remember). The problem is not service disruptive, so I don’t check on it that often. A family member was streaming remotely around that time, so I would have heard if the connection dropped.
The steps I mentioned in my post were at ~11:00 today, 27 April.
I power cycled my router at ~09:30 on 27 April for an unrelated reason. You may see that in the logs as well.
I’ll shut down Tautulli and see if it makes a difference. It is a “nice to have” for me, not a “must have.”
I will have a debug build with extra logging that I would like to make available to you to help locate the bug that is causing remote access state to get unset. It is based on 1.19.2 and a good test would be to have it running with the 1 minute Tautulli tests as that would spot the issue quicker
If you are happy to run with, please let me know what binary file you need for your platform
I too would be interested to know if the new build fixes the problem, as I have the same system (Synology DS918+ Intel 64-bit) and I am also experiencing the same remote access disconnect problem.
Whilst your logs showed what looked like a different issue from what was just fixed in beta 1.19.2.2737, I would like you to try it and let me know if there is any difference. May be the two issues together made the problem more severe than just what I observed in your logs
If the problem persists in 1.19.2.2737 please get me server logs for the time with both verbose and debug enabled
There were two causes for this problem - one, a regression in early 1.19.2 betas was fixed in 1.19.2.2737 and the other is awaiting a change to a timeout period on the back-end plex.tv systems that do the remote access connection testing.
A Plex Media Server log with verbose logging enabled and covering the time when a connectivity test is done would confirm if the issue is down to the timeout problem
I have the same diffulty that you describe. Sometimes my Plex will run for 8 to 10 hours but eventually (without me doing anything) it resorts back to red and my remote access will not work.
It sounds like you have a different problem. The issue on this thread was that remote access still works, even when the Remote Access config page said it was not available (a false negative). It appears resolved with the PMS 1.19.2.2737 release and a fix on the plex.tv backend.
Suggest you open up a new thread, so it will get the necessary visibility (instead of this 10+ day old thread). Update to PMS 1.19.2.2737 or later to eliminate the false negative issue. Set your server for debug logs. After the problem occurs, pull the logs and attach the zip file to the new thread (the logs need to capture the transition from available to not available).
I’m running 1.19.2.2737 and the issue did disappear for a while (can’t say how long, maybe a day or two) but its back again. External users can connect fine but Remote Access dashboard reports it as offline.
I will look at your issue once you are on the current release of Plex Media Server as there have been changes between 1.19.2 and 1.19.4 and there are also better diagnostics within the serrver logs
Briefly switched to available then to unavailable after jumping from ‘Remote Access’ page to ‘General’ and back to ‘Remote Access’ again after upgrading to Version 1.19.4.2935
As before external devices connect and stream fine.
That appears to be the Australia / New Zealand issue where we are aborting / timing out on the connectivity test request - the timeout is supposed to be 5 seconds
Jun 30, 2020 23:00:12.875 [5160] VERBOSE - Didn't receive a request from 34.243.160.122:51686: stream truncated
Jun 30, 2020 23:00:13.152 [2176] DEBUG - EventSource: Got event [data] '<Message address="118.211.xxx.xxx" port="45981" asyncIdentifier="2d12b04d-bc7b-4323-a561-6a3b9ab3faea" connectivity="0" command="notifyConnectivity"/>'
I will let our operations team know that this is still an issue for some servers far from the Ireland based Amazon Cloud servers that do the connectivity tests
