No Remote ACCESS / DLNA not visible on LAN - until server gets restarted

Hey,

I am having issues with the DLNA over the LAN. Plex media server is not visible on the LAN. Once restarted server becomes visible for some time, then again cant connect after some time passes.

I migrated existing setup under dockerized setup and same scenario. I google around that it could be some conflicts but all required ports were acquired by Plex service…

However recently I was forced to do the clean setup of the server and tested it again with the same exact behavior…

root@caradhras:[~]: @netstat | egrep "1900|32469|3005|5353|8324|32410|32412|32413|32414"
tcp        0      0 0.0.0.0:32469           0.0.0.0:*               LISTEN      1006       99769      11812/Plex DLNA Ser
udp        0      0 0.0.0.0:32410           0.0.0.0:*                           1006       97393      4012/Plex Media Ser
udp        0      0 0.0.0.0:32412           0.0.0.0:*                           1006       91050      4012/Plex Media Ser
udp        0      0 0.0.0.0:32413           0.0.0.0:*                           1006       1662596    4012/Plex Media Ser
udp        0      0 0.0.0.0:32414           0.0.0.0:*                           1006       97386      4012/Plex Media Ser
udp        0      0 0.0.0.0:1900            0.0.0.0:*                           1006       112212     11812/Plex DLNA Ser
udp        0      0 0.0.0.0:5353            0.0.0.0:*                           109        22008      805/avahi-daemon: r
udp6       0      0 :::5353                 :::*                                109        22009      805/avahi-daemon: r

Any ideas? Plex server not visible until I execute docker restart plex. Then right after I can see it in TV… To me it looks like your internal bug :confused:

Much appreciated. Thanks, Michal

Server Version#: Version 3.77.4
Player Version#: DLNA in Sony Bravia or VLC player at PC

Sounds like a subnet or firewall issue.
Docker creates a whole host of issues with networking UNLESS you configure it to use –host networking. It sits in the container, in its own little networking bubble.

I am not a docker person but if you make the container use the host’s network resources, it will become visible.

Sorry but no… This doesn’t sound like subnet or firewall issue at all. Why?:

  • It’s LAN, so firewall doesn’t take place
  • Servers internal firewall is allowing the traffic to pass doesn’t it? Anyway tested it with disabling it and didn’t help.
  • When you talk about the subnet issue you need to understand at which layer of the OSI model you operate. Since both the server and the client are on the same network then that is the 2nd layer of the OSI model and simply there is nothing which could interfere with the connection like this. Simply it works or not, simple as that.

Secondly, no need to explain docker and be concerned about it. Pointed out that this was happening before I made my Plex setup dockerized. Anyway the Plex service is put directly on the host network as you suggested. However you cloud understand that from the output I provided :wink:

Anyway this issue was reported here even before but nobody seriously looked into it yet.

I explained the process of replicating this problem and it doesn’t seem that hard to isolate the scope.

If we agree on the conditions that neither network, firewall or any other host configuration changes then a) Plex DLNA server is not visible, we b) restart it, then every time c) Plex DLNA becomes visible proves there is an application layer problem. So with the Plex media server itself.

Hope now we can understand each other better and will avoid further misleading discussion…

Thanks, Michal

Can you provide some additional information?
What is the Host OS?
If you stop the plex docker and run your netstat on the host, is port 1900 already taken?
3.77.4 is the Media Player version not the Server version.
Do you have Vlans?

Hi,

Sorry my bad, server version is: 1.14.1.5488.

Host system is Debian Buster, before it was Debian Stretch (9.5).

root@caradhras:[~]: cat /etc/os-release
PRETTY_NAME="Debian GNU/Linux buster/sid"
NAME="Debian GNU/Linux"
ID=debian
HOME_URL="https://www.debian.org/"
SUPPORT_URL="https://www.debian.org/support"
BUG_REPORT_URL="https://bugs.debian.org/"

No vlans on the network.

After Plex was stopped:

root@caradhras:[/docker/plex]: @netstat | egrep "1900|32469|3005|5353|8324|32410|32412|32413|32414"
udp        0      0 0.0.0.0:5353            0.0.0.0:*                           109        23517      893/avahi-daemon: r
udp6       0      0 :::5353                 :::*                                109        23518      893/avahi-daemon: r

Thanks a lot, Michal

EDIT (BTW):

root@caradhras:[/docker/plex]: type @netstat
@netstat is aliased to `netstat -tupanve'
1 Like

Hi, any progress? I can help with whatever you need to help this resolved once for all :slight_smile: .

Turn off IPV6 support in Plex networking. Any reason you are relying on DLNA instead of using one of the Plex Media Player clients?

If this is still causing problems follow the steps to upload the Debug(not verbose) logs:

https://support.plex.tv/articles/200250417-plex-media-server-log-files/

I will try turning IPv6 off once I get home.

Reason is simple. It is way faster than via Plex clients in TVs… Although I have powerful server clients in TVs tends to get stucked and are laggy. While using build in player for the DLNA the experience is way faster and realtime. I have Samsung, LG and Sony Bravia TV and in all clients the experience via Plex client is far from ideal. The worst is for Sony Bravia where users are force to run Plex client through Opera store…

IPv6 support was already disabled.

Here are the logs:
Plex Media Server Logs_2019-02-05_14-36-19.zip (5.6 MB)
Thanks

You can turn off the verbose logging.

From what I can see, it can’t determine it’s public address and after a time it can’t figure out what the gateway device is so it stops sending. This could be an incompatibility between your router and Plex with regard to UPNP, router dropping the port assignments or you are doing additional layers like VPN or reverse proxy and the gateway device is changing. Since this problem existed before you dockerized you may have to go back to basics and start simple adding layers one at a time.

Sorry for late response… I just did serious debuging of my network… And it seems the problem has bigger scope than just LAN.

The whole problem:

But my devices connect from WAN. However I remember having my server couple times “offline” while I was sure there was no connectivity issue.

My Router: Ubiquity EdgeRouter PoE v1.10.8 (upgraded to latest image)

At the time I wrote this topic I did not have UPnP set up at my router. After suggesting it I enabled it via GUI - no change. Recently I started having complains from my sister in different city that Plex is unreachable from time to time… I didnt notice because I access it 95% of a time from LAN. So I started playing with the network setup.

My default configuration is so called static port forwarding. Which works for all my services except plex.

  1. Static portforwarding (999->32400): with or without Hairpin NAT (NAT loopback or NAT reflection)
port-forward {
     auto-firewall enable
     hairpin-nat enable
     lan-interface switch0
     rule 1 {
         description SSH
         forward-to {
             address 192.168.255.11
         }
         original-port 22
         protocol tcp
     }
     rule 19 {
         description Plex
         forward-to {
             address 192.168.255.11
             port 32400
         }
         original-port 999
         protocol tcp_udp
     }
     wan-interface eth0
 }
  1. Destination NAT
  1. No UPnP
  2. UPnP
     upnp {
         listen-on switch0 {
             outbound-interface eth0
         }
     }
  1. UPnPv2
    upnp2 {
        listen-on switch0
        nat-pmp enable
        secure-mode enable
        wan eth0
    }

None of the above fixed the “Remote Access” problem, neither DLNA on LAN…

Every time I did a configuration change on the network I restarted the docker container for plex service, then tried to re-enable remote access again. And the results is, that for like 2-3 seconds it blinks green " Fully accessible outside your network"

but right after that it goes again:

I am really clueless now and desperate to fix it :confused: Anyway attaching again my logs.
Plex Media Server Logs_2019-02-21_21-42-46.zip (3.5 MB)

Appreciate your help

Reread everything and I see I have complicated the issue. I didn’t see any issues in your log to indicate network problems.

At the moment, Plex is known to not properly show remote access correctly. Port forward with UPNP turned off is best. As long as remote clients are working then all is ok, this is separate from DLNA.

DLNA only works on the local LAN, which gets back to the Docker config, Host mode is best so that DLNA doesn’t see 192.x.x.x as a different network from 172.x.x.x. Docs on the DLNA settings are a bit vague. Maybe AVAHI would help but I don’t know, I have it running on my router and Debian host. I’ll install the avahi-utils and explore a bit.

And there is a new server version to check out.

Thanks @pl_5309 for your effort and energy… I just finished experimenting this problem with my old TP-Link using OpenWRT… I used it before and recall no similar issues. However when I replaced it, same behavior… I noticed I didnt have any UPnP package installed (and it used to work)… This time I checked and installed one to test.

And exactly same scenario… Remote access enabled for short period of time and then again “not available outside your network”…

dokcer-compose.yml

  plex:
    image: plexinc/pms-docker:plexpass
    container_name: plex.529
    restart: always
    network_mode: host
    volumes:
      - $PWD/plex/config:/config
      - $PWD/plex/transcode:/transcode
      - /lvm/raid/private/media:/plex/media
      - /lvm/raid/private/data:/plex/data
    environment:
      - TZ=Europe/Prague
      - PLEX_UID=1006
      - PLEX_GUID=1012
******* just added *******
      - ADVERTISE_IP=http://192.168.255.11:32400/
      - ALLOWED_NETWORKS=192.168.255.0/24,10.8.0.0/24
**************************
    labels:
      - com.ouroboros.enable=true
      - traefik.enable=true
      - traefik.port=32400
      - traefik.frontend.entryPoints=https
      - traefik.frontend.rule=Host:plex.example.com
      - traefik.backend=plex.529
      - traefik.frontend.headers.forceSTSHeader=true
      - traefik.frontend.headers.STSSeconds=315360000
      - traefik.frontend.headers.STSIncludeSubdomains=true
      - traefik.frontend.headers.STSPreload=true

Regarding DLNA, as I mentioned, Plex server was always on the host network. Anyway in the meantime I’ll try to explore avahi-utils and let you know.

Is there any chance some dev guy might look into it?? Thnaks

EDIT:
I updated a little bit the docker-compose file per official image instructions and again same scenario.

EDIT2:
Just what I have observed for the past 2 days - no changes to the network and DLNA sometimes works, sometimes not…

Any suggestions?? Remote access problem still happening and DLNA not working. I am willing to assist with troubleshooting.

What am I supposed to do to get some attention please?

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.