Plex not connecting to home network clients (or indirectly). Only remote access/externally

Server Version#:latest; 1.43.1.10611
Player Version#: various different devices

OS: Win 11 Pro

Started: 5-6 days ago

ISP: Bell, modem: Gigahub 2.0.

Server/Plex Connection: Ethernet

Started maybe 5-6 days ago, randomly. Before that, everything was working fine for about 12-16 months. Plex Web UI and access works fine on the host server PC. Works fine with remote access, off home network, or on data, etc, However any other device, PC, or client trying to connect on the same network fails (without relay) or is indirectly connected (with relay).

I’ve been trying to troubleshoot all week, with no luck. I’ve ensured I’m not in a double NAT situation with my modem, although I do have TP Link working as an AP for my modem, and all is working fine so I can tell. I’ve restarted my modem, along with my server PC, several times now, I’ve rolled back to three updates ago (the oldest update Plex had for me in my DB), and then updated to latest again, along with other troubleshooting methods (and even the help of AI chat, which was completely useless), and still no dice.

I’m happy to post the logs here in this post if I need to, since I’m out of things to do. Just wanted to make sure what to do right.

If there is any other information needed, I’m happy to give anything I need.

Thank you

<If providing server logs please do NOT turn on verbose logging, only debug logging should be enabled>

  1. Important note about the Plex Web app - Local Network Access
  2. https://support.plex.tv/articles/206225077-how-to-use-secure-server-connections/
  3. https://forums.plex.tv/t/client-says-it-wont-play-from-remote-server-although-the-server-is-local/921148/1

Hello.

I use Firefox, and that setting has always been on on my end, so that wasn’t the issue. I did come across that first & second post and tried, with no luck.

Also to add, it’s not just the web UI that I’m having issues with, it’s absolutely ANY client or service connecting to Plex under my home network. Isn’t browser related at all, since my phone’s app, TVs app, tablets app, and even kodi Plex add-on also weren’t working, but then work fine once I connect to my remote mobile hotspot.

That setting doesn’t exist yet on regular releases Firefox, so it couldn’t have been on for a long time.

#2 is very relevant, as it deals with DNS resolution in your local network
#3 almost everything in there can be relevant

When I click the symbol next to the link on my Firefox browser, it comes up with a pop-up that says “access local network devices: Allowed”, which is what I was referring to. If that’s not the exact setting you meant, then apologies. However it seemed to me like it was, or filled similar content.

As far as #2, Idk I’ve never had issues with any DNS from my modem or ISP for just about 2 years now. Would this be the kind of thing that would just suddenly and randomly happen? I’ve attempted changing my DNS on the server PC to the CloudFlare or Google 8.8.8.8 and 1.1.1.1, changing it from automatic to manual DNS settings in Windows, restarted both the machine and service, and still no luck.

And for #3, I’ve tried some of the troubleshooting suggestions in it, but not all so I’ll give the ones I haven’t a shot, but I’m not optimistic considering the extent I’ve looked into this now. I guess I was hoping for more guided support, or assisted guidance.

Do you mind if I go through the solutions in #3 and address each one that doesn’t give me a solution?

  1. Im not using a container for Plex. It is directly installed on my PC. Although I do run Docker Desktop in the background.

  2. I dont know of the many ways to confirm this or not, but my server PC is on a static IP that has been reserved for it since I first set it up. I’ve also tried loading Plex through https://192.XXX.XXX.XXX:32400/web, and it just loads forever until the page errors out. And as I mentioned in the previous post, I’ve tried setting my DNS on my PC from auto to manual and tried different DNS (cloudeflare and google), and restarted with again, no luck.

  3. There are none. Only one modem, and attached to it on a different floor, one TP link access point device, which I’ve both restarted and turned off during some troubleshooting to verify if it was an issue.

  4. My home network devices have private IPs that read 192.XXX.XXX.XXX, with the same public IP, as verified with the Whats My IP sites, and device settings.

  5. None. No add-ons, and isnt an exclusively browser issue.

  6. Tried. It loads and loads, then errors. Error is boilerplate: (The connection has timed out

    The server at 192.XXX.XXX.XXX:32400 is taking too long to respond)

  7. They do not. not exclusive to browser.

  8. Not applicable. Windows

  9. Have Private Internet Access and Tailscale on the PC, with split tunneling enabled, however Ive tried both disabling them and troubelshooting, and enabling them and troubeshooting, along with disabling them from network connection settings completely, and troubleshooting. No luck.

  10. I don’t fully understand this one, but remote access port is, and has always been, 32400 and opened as such through my modem. Also, remote access it not an issue at all. Everything is working completely fine on that side.

  11. It was originally set to Public (I believe by default, or as recommended by windows), but switching to Private, restarting, switching to public again, restarting, all provided no change.

If you try to access the server from another device on your network using the direct ip:port in a browser does it also time out?

I was trying to help another person on Reddit that’s also on Bell using the Gigahub (they originally said Roger’s, but corrected it later) and no device on their local network can access the server, even in a browser using the direct ip. The port shows as open when you test it locally via terminal, but nothing can connect. They run their server on a NAS, so they can’t even connect to the sever to manage it. They can reach the NAS, it’s just port 32400 that won’t resolve.

They have some other custom apps running on their network that they plan to disable and test again..but just wanted to share their issue:

I saw some other posts recently from users on another ISP, YouFibre, where the isp router was blocking plex locally. Not rebind or isolation issues…Same behavior that this user was experiencing…the local server ip and port was being blocked. ISP provided a different router or users used their own and the issue went away.

Yes it does also timeout. That was one of the methods that I came across, and for reference it would look like “http://192.xxx.xxx.xxx:32400”.

For experimenting after reading what you mentioned, I decided to open port 8989 for Sonarr to see if it can be accessed, confirmed it was open, and although the connection is very slow (glacially slow), it did connect to the Sonarr log-in screen, and is currently loading me into the service, albeit has not fully loaded yet. (edit: and did NOT load me in. no error, but am met with a blank page. Although I was initially met with the auth screen)

After reading the other Reddit post, I guess something to add, I also don’t have remote access setup on Jellyfin for instance, however it did manage to work on my devices as a fallback to when I couldn’t figure this out with Plex, with no issue. Obviously I’m not aware of the differences of how both apps handle what, but just to note.

You need to use http, not https. The local web app cannot be loaded securely over https.

Are you certain that
a) the modem is just that and not a full featured “router” instead
b) the access point is just an access point and no a full router device
? You should have only one real router in your network.

So even two VPNs at once? Many, many times it has been reported in here that even one VPN can wreak havoc.

Definitely. ISPs are messing with DNS resolution all the time. Either to purportedly “protect” their customers from harmful websites, or to earn additional money or suppress websites and services which may causing them to lose money. (like, for instance a privately run video streaming server, which reduces the customer’s inclination to use vido streaming services with which the ISP may have revenue sharing arrangements.)

A static IP for the Plex server device is definitely a good thing and is very much recommended.
However, it has nothing to do with the issue of DNS rebinding protection being applied in your local network. That is done by your local DNS resolver, which is typically situated in your router. So you need to look the the settings menu of your router to find the relevant preferences (if there are any available, that is).

I dont know why I wrote https (maybe force of habit), but it was http, along with my experimented connection to sonarr. My mistake.

The gighub 2.0 is described by Bell as a “modem-router”. As I had poor wireless signal on the top floor, I decided to invest in a TP link that acts as an AP for it. I’ve confirmed through both the TPLINK settings and usage that it is acting as an AP, and not as an extender or some such. I have also tried troubleshooting with it off, and to top it off, my main PC that is connected via ethernet is having the same reported issue, not just wireless (which is what the TPlinks AP is for).

Yes, I have PIA with split tunneling (and also have everything except for my torrent client meant to bypass the VPN) for both the port (32400) and the app Plex, and have tried with it enabled and disabled, and the issue persists. I can try uninstalling it and trying again, but something tells me that wont solve the issue since I went into Connection settings and manually disabled it to try. as for Tailscale, its only active and on when in use for remote gaming, and is not on at all times. However I’ve had both of these services setup on my PC for almost the same exact time period as Ive had Plex setup on it, with no shadow of an issue.

Unfortunately on my modem/router, there are no DNS settings I can dive into. Just a section telling me what my current IPv4, IPv6 DNS Primary and Secondary DNS are, and a space for manual specific DNS info input. As for DDNS, that is disabled.

This should be set to “private” and remain so. Setting it to “public” will suppress any attempts to contact the server from within the local network.
Windows will revert it automatically to “public” if it detects a change of the network, like e.g. a different router etc.

Same problem here and I found out why but not how to fix it.

I am on Bell with a GigaHub 2.0 modem with 1.5 Gbit fiber. I factory reset the modem/router and turned off my Pihole.

Magic! I could log into Plex and see my media. Then I went into settings and noticed that remote access was not working. I had forgot to set my port forwarding on the modem/router. I setup port forwarding and now I cannot access Plex locally anymore. If I turn off port forwarding I can access Plex locally.

I noticed the IPv4 only note on the port forwarding page. I am going to see if I can connect to plex locally using IPv6. I now just have to figure out how to do that :slight_smile:

So…it seems the problem is port forwarding.

Try turning off port forwarding and see if you can access Plex locally. That fixed it for me. I think I am going to have to have 2 servers running. One for local access and one for remote access.

PS. I was trying to resolve this on Reddit but I was pointed to your forum post and will try to work through it here with you.

>This should be set to “private” and remain so. Setting it to “public” will suppress any attempts to contact the server from within the local network.
Windows will revert it automatically to “public” if it detects a change of the network, like e.g. a different router etc.

Duly noted. Ive set it to private and will make sure it stays private, although no fix.

@cfus98

I mean, I can’t believe it but you’re right’ I deleted the port forwarding rule in my modem, and it connects locally (although shows remote access with a check, even though I tried connecting externally and it connects indirectly, obviously. Idk if that just takes time to update or not)

I’m at a loss, and I also can’t make heads or tails of this situation. So turning on port forwarding to allow it access outside your network, causes it to lose all connection inside your network. and then when its turned off, it gives you access in-house again… I have no idea where to go from here now hahaha

( I was also directed by a mod here to your reddit post, and it was “relieving” (thats one way to describe it I guess lol) that someone else was having this issue, and it wasn’t just an anomaly on my end.)

Glad you guys connected, hopefully you can figure it out. Very odd behavior you’re experiencing…

Just fyi..I’m not a mod.

Does running two different servers side-by-side seem efficient, though? is that not two completely different instances of Plex? Idk that I can do that, at least personally. I’m struggling to figure out where the issue could be coming from to have something like this occur. It’s not even like any settings, updates, or changes were made within the past week or two, and this issue started for me right around the same time yours did.

Could it be a modem issue, hardware wise?

(@veeejay thank you very much for the connection. Didn’t realize you weren’t a mod, but you helped in bringing light to the issue, so you are a mod in my heart lol)

I am using an Unraid server and have Plex in a docker container. Here is the setup I am testing out and it appears to be working with port forwarding turned on in the Bell modem/router. Server is Gen8 12core with 48GB ram and 3080 nvidia card.

Container 1: Official Plex docker container (plexinc/pms-docker). I am using my claimID on this server so I can use remote access and transcoding to devices like my tablet. The network is using Host. Not accessible locally but I can access is outside my local network.

Container 2: Linuxserver Plex container (linuxserver/plex). No claimID so no transcoding. I pointed it to the same media files and a new metadata library is being built right now. The network is Bridge and I gave it a fixed IP. It is not accessible outside my local network but I can access it locally.

Container 3: Official Jellyfin container (jellyfin/jellyfin). This is my backup media server for when I lose internet. It is also going to be the server if I need to transcode inside my local network.

I also run JellyPlex Watched (luigi311/jellyplex-watched) to keep all the media server watch lists synced.

Yes, this is overkill but sometimes you just want things to work. I will shut down the containers if and when the port forwarding starts working again.

Hhahah thanks!! Just don’t want to take away from the hard work and time the mods and employees put in to supporting us.

I personally think this is a hw issue with the bell router. Given users on another isp experienced something similar and even the isp confirmed it and replaced the router with another model to fix it, shows maybe some firmware update broke something. Can you bridge the router to use your own? Or bypass it entirely?

Appreciated. I am currently running a Win 11 pro on i5 10600K, 32gb RAM and a 1070 Nvidia card. Plex running locally installed. I have docker desktop running with services, I suppose I could run a second Plex instance through docker, but Idk that I have the memory to have it rebuild a new database, as my nvme OS is 1TB and nearly full, with about ~100gb to spare (and idk how to redirect plex to store its DB on my HDDs).

Sorry, but can you briefly explain to me the claimID and networking bridge bit? I got kind of lost on that second container reference line.

@veeejay similarly with my aforementioned question, can you briefly explain bridging the router to me? I will admit I am kind of intermediate when it comes to network specified details.

Bell only allows bridge mode for buisnss clients. Our only option is Advanced DMZ.