Will do. Looking into it now.
Looking into the networking problem I found that Windows broke on the back end. Not sure what broke exactly, but after a reset of the Server PC everything worked, local and remotely. I backed up and rebuilt my Plex server before writing this post to ensure all repairs were good.
I want to note some of my findings while troubleshooting
I found Asus routers do have DNS Rebind Protection enabled and is not able to be turned off within the stock firmware. I highly recommend installing the Merlin firmware for Asus routers for the ability to turn off DNS Rebind Protection.
Check your firewall rules within Windows. Apparently Windows updates will change them.
Backup your database. Plex has an article about how to backup and reinstall your data. This has all your watch data and viewstats.
I reset Windows after this. After the reset and setup of Windows I reconfigured Plex and found everything was working. Reinstalled my database backup and metadata to a fresh working Plex.
Massive thank you to @sa2000 for your patience, quick responses and troubleshooting skills with this matter. I could not have done this without your help.
Problem solved.
Thanks for the feedback. Glad that the issue has been resolved.
Hello!
I have the same issue.
My configuration is basic : the server (v 1.19.4.2935) is on Ubuntu 20.04 with static private IP on my LAN, firewall disabled, the 32400 port forwarding is set on router.
It works fine on LAN.
When I perform the test of remote access, everything goes green for a few minutes then turns red. Itās impossible from outside my LAN to connect to the plex server without relays (from web browser or from phone app).
Is there a plan for fixing this issue?
Thank you!
which issue ? every case is different and the majority of actual issues are down to user configuration / environment / ISP -
The Troubleshooting Remote Access support articles gives a number of causes
Some common themes
- Double NAT
- CG NAT
- xFinity xFi Advanced Security
https://support.plex.tv/articles/200931138-troubleshooting-remote-access/
So I think I fixed my issue that has been bugging me for a few months.
Plex on unraid.
Plex manually port forwarded to 32400
PFsense router running pfblockerng.
Remote access is fully accessible when manually tested, but then fails several minutes later.
Here is the thing, I do not believe subsequent checks for connectivity originate from my plex server. Instead, an amazon aws ip address attempts to hit my server on the forwarded port 32400. Today at least that server was located in western Europe. I geo block all of Europe.
So I followed the advice https://forum.netgate.com/topic/139324/pfblockerng-permit-inbound-wildcard-domain/2 to add allow rules for AWS servers. I did have to add a couple advanced inbound rules for destination port and protocol.
Remote access has been holding steady for an hour or so now.
The issue where it works but doesnāt display a green tick on the server. If I disable secure connections and change nothing else, I get the green tick.
That is not the issue I was responding to @FafTux
Itās impossible from outside my LAN to connect to the plex server without relay
Your problem could be the timeouts on connectivity testing i have seen on servers in Australia and New Zealand and that may still be there. We have a 5 second timeout which should be sufficient. As far as I can see, there is no real impact other than the status indicator.
This issue bugged me for ages, my resolution was to manually set a new LAN IP for server and make it static. With this in mind I introduced a DNS I have not used before as 1.1.1.1,1.0.0.1
I live in Australia and since applying the above, remote access has been flawless. Happy Days
I have this set at Server, Gateway Levels and to avoid rebinding issues have it applied to my VPN. Checks have proven no leaks.
I use a split Tunnel VPN with application exceptions for all things PLEX and Australian streaming services that have Geo restrictions with a dedicated Browser.
Iām in Australia so this could be the issue. I donāt see any other problems but a five second timeout would indicate a problem exists.
and the countless hours spent trying to make it work.
@SE56, I donāt use a VPN on the Plex server and Iāve tried several different DNS servers but 1.1.1.1 hasnāt resolved it either. Thanks for the suggestion. The serverās IP is static on the LAN.
You must restart everything after applying to make sure it has taken
@sa2000 I want to share my solution after reading the 250 comments about this issue. I am from Argentina, Fibertel internet Service with SAGECOM FAST3686 Cablemodem. The remote access connected for a few seconds and then it dropped after I moved Plex Server due to a windows crash.
The solution: MODEM ROUTER COMPLET HARD RESET (with tiny reset Button)ā¦when Modem went online again, I start Plex Server and the magic was done. (Before reset everything looks right on portforwarding upnp, but not working). Server has Fixed IP Address. Sorry my English, and thanks to everyone of you, It was very helpfull and instructive reading all your posts. SORRY MY ENGLISH!!! we speak spanish down here.
Thanks for the suggestion @fcaniete. I tried the full factory reset on my ASUS RT-AC88U and it almost - but not quite - nearly appeared to work. I was seeing the green tick of eternal happiness for as long as I left the remote connection tab open, even overnight. As soon as I select another tab like Network, back to red. I can browse my library and then back to settings and it stays green as long as I donāt switch tabs within the settings. This is different behaviour to what I have been seeing and led me to some further conclusions.
The difference was UPnP on by default. Turning this off and relying on port forwarding causes the green tick to disappear a few seconds after appearing.
Further to this, using 32400 as the external port the same as the internal one also maintains the green tick for as long as I stay on the remote connection tab. This also requires the manual port option to be unticked.
In summary:
In all the following scenarios, canyouseeme.org always reports success and
remote connections work just fine. Itās just the red status of eternal annoyance.
Disable secure connections - green tick forever regardless of selected tab.
UPnP on - green tick as long as I stay on the remote connection tab.
UPnP off and 32400 as the external port - green tick for as long as I stay on the remote connection tab.
UPnP off, 32400 as the external port with manually specify public port ticked and 32400 as the port - green tick for a few seconds and then red. No need to switch tabs.
UPnP off, any other external port and manually specify public port ticked - green tick for a few seconds and then red. No need to switch tabs. This is what Iāve seen the most over the last few months because this is the scenario Iād like to use and always have used in the past.
I have no idea what Iām doing with Wireshark but it looks like this when it fails for two of the Plex IP addresses listed above. The other Plex IP addresses have zero traffic.
Back in your court Plex and thanks for fixing all the Shield issues with freezing and forgetting what weāve watched that plagued us through lock down. Iām lucky to know Plex normally just works as it has done for many years. For newer users, itās well worth hanging in there.
The issue appears to be what has been seen on other servers located in Australia, New Zealand, Malaysia, Singapore.
A change is being worked on now to double the timeouts for connectivity tests that get sent from the Amazon cloud Plex servers based in Ireland to servers in those identified regions with aborted connectivity test requests.
This is consistent with what has been seen. The https TLS dialogue adds delays and the http insecure requests are much faster
If a specific public/external/wan port is to be used then the following must be in place:
- static and/or dhcp reserved IP for the local IP of the server
- a port forward of that public tcp port to local port 32400 to go to the local IP of the server
- manually specify port must be ticked on the server remote access settings page and the public port number entered in the box
- uPnP can be enabled or disabled - it is better if enabled as it is used to find the public IP address and compare it with what plex.tv sees to determine if there is a Double NAT
If a dynamic/random/automatically allocated public/external/wan port is to be used then the following must be in place:
- uPnP must be enabled
- Manually specify port unticked in server settings for remote access
In all cases, sometimes it helps if there is a firewall rule on the server computer, allowing incoming connections to local port 32400 from any remote IP and any source port
Thanks for the detailed reply @sa2000. I know you help a lot of people with remote access. Iāve read a lot of threads on the subject this year and feel I have a handle on how it all fits together. You have incredible patience dealing with the many different causes.
In all the scenarios I listed, configuration is correct as indicated by the green tick whenever secure connections is disabled. Plex behaves a little differently depending on the chosen configuration, which is why I thought Iād document it.
It sounds like Iām suffering from the TLS timeout issue and should have ignored it for six months but itās hard to ignore an issue for so long without any official feedback. Iāve wasted a lot of time on it and Iām sure others have as well. The only way of knowing about the suspected problem is to trawl though forums. On May 5th, you said the TLS issue had been resolved.
Someone should add a sticky with a description of the problem, the symptoms, and an indication youāre still working on it.
@heapson
@kesawi
@gandalfarcher
The timeout periods have just been doubled for servers in your countries. I would like to know if the issue remains and if the requests for connectivity testing are still aborting, I would like to get wireshark capture and associated Plex Media Server tls key logging. I would send you instructions how to get that in Private Message if the issue remains.
Thanks for trying @sa2000. Unfortunately nothing has changed at my end. I havenāt retested all scenarios but the one I use, a fixed external port forwarded to 32400 still shows a status of red after a few seconds of green.
Disabling secure connections and changing nothing else gives the green tick suggesting itās configured correctly. With secure connections set to required, remote access works fine on my phone using 4G and displays āRemoteā in the dashboard.
I tried a different browser without any privacy extensions but it was the same. My router is currently running the latest ASUS firmware with almost default settings and all non-default changes documented but Iāve tried the Merlin firmware release as well.
I donāt have time right now but in a few days, Iāll configure a clean vanilla Windows 10 dual boot installation with just Plex and Wireshark and PM whatever logs you require. This should remove all the potential software interference. Iāve done this before and saw the same behaviour.
@sa2000
I too am again having this issue, I fixed it until you guys ran another update 
, I just added a few tweaks in the router and all was right with the world, Remote was stable but then the update took away my glory! I have all my Data tell me where youād like it sent please and off itāll be.
TIA
Beefā¦
@sa2000, apologies for delay but have sent logs and packet captures as requested in PM. Remote access indicator is a bit more stable, but still changes from green to red intermittently. Logs should contain instances where this has occurred.
got nothing?
