Just wondering why plex doesn’t support more than 1 direct play session? Both my brother and a friend use the Nvidia Shield keeping their plex players and the Shield up to date, when I see one of them on, it’s using direct play, when the other gets on it start transcoding (hw). If the other jumps off the transcoding (hw) switches to direct play.
My Plex runs on a Dell R810 with ESXi 6.7 and a Nvidia Quadro P2000. The Plex server has 12 cores, 32GB of RAM, SSD transcoding drive, the P2000, etc.
Same thing happens when I get on my FireTV 4k Sticks or Cube 2019 while my friend and brother are on, my FireTV grabs the direct play while their system switch to transcoding (hw).
Did you impose any limits on the internet connection? It might just be that one DP session is allocating so much bandwidth, that a second DP session cannot be started. Because the remaining upload bandwidth is not high enough to accommodate the peak bandwidth of the second file, additionally to the first DP stream.
There is more then enough bandwidth on my upstream to support 2 to 3 people doing direct play. I don’t limit my connection.
And you didn’t explain why when I watching stuff locally over my Gbit LAN the direct play switches over to my FireTV and disables for my brother and friend? Clearly that’s not a bandwidth issue, I could see if my devices were going over the internet but they aren’t. I have my LAN networks setup in my Network settings.
The rest of my house consists of two CISCO 3560G 24/48 port POE with CAT6 and CAT6A shielded cabling. And my Ubiquiti NanoHD’s.
I don’t know how big your upstream is.
“more than enough” is hardly quantifiable.
Inspect the Plex XML info of the two items which were played simultaneously.
Seek for the first occurence of the requiredBandwidths="... property in each.
Look at the first number in there.
This will tell you the actual bandwidth requirement to Direct Play this file. These are kbps.
Add these numbers for the two different files. Now add another 20% safety margin on top. How close is this sum to the nominal upstream capacity of your internet connection?
Your Plex server is not the only device using your internet connection. How many other devices are connected to the same router?
The backbone of your internet connection might share its capacity with your neighbourhood. The actually usable upstream bandwidth can get much lower than the nominal number at certain times of day.
Are you sure that your local Plex clients are actually seen by your server as “local”?
Are you using a fragmented local network with several “router” devices?
Is your local network using a “private” IP range? https://en.wikipedia.org/wiki/Private_network
Nominal 50 mbps is not that much. I bet the actually usable bandwidth is often below the 50 mpbs mark.
I technically get between 51 and 54Mb on my upload speed. I just rounded it down to 50Mbps. Again this is business service internet, not hokie pokie home service.
There is nothing special about my home lab. I don’t have my IOT devices on separate VLANs. Can you tell me were in the plex server I can see if the device is being seen a local or external device? I just had a quick glance and don’t see anything in the Windows server that shows this.
Yes, my network setting is using my private IP/24.
activate debug logging (not ‘verbose’!)
2. quit Plex Server
3. wait 1 minute
4. start Plex Server
5. wait 2 minutes
6. play the movie, take a note of the time when you press Play
7. wait 3 minutes
8. fetch log files
Inspect the Plex Media Server.log. around the timestamp of when you pressed play. You should see the IP of your client device and in square brackets either [WAN] or [local] or [nearby]. If it says [WAN], it is seen as a remote device.
Make sure that your server VM is using the host network and is not NAT’ed.
I did what you asked, accept I used my Samsung phone to connect which is connected to my WiFi at home on the same private IP address network as the Plex server. When I check the debug log I see this… (edited some stuff out)
Nov 20, 2020 08:59:08.797 [10548] DEBUG - Request: [135.xxx.xxx.xxx:43662 (WAN)] GET /video/:/transcode/universal/session/02111c36369e7b68-com-plexapp-android/base/00120.ts (28 live) TLS Signed-in
Nov 20, 2020 08:59:08.797 [10548] DEBUG - [Transcode/02111c36369e7b68-com-plexapp-android] Asked for segment 120 from session.
Nov 20, 2020 08:59:08.797 [10548] DEBUG - [Transcode/02111c36369e7b68-com-plexapp-android] Returning segment 120 from session
Nov 20, 2020 08:59:08.798 [10548] DEBUG - Content-Length of G:\Transcoding\Transcode\Sessions\plex-transcode-02111c36369e7b68-com-plexapp-android-e44480bd-8bed-4ca2-8a25-1bd13de25b49\media-00120.ts is 2997444 (of total: 2997444).
Nov 20, 2020 08:59:08.927 [10548] DEBUG - [Transcode] Session 02111c36369e7b68-com-plexapp-android (3) is unthrottling
Nov 20, 2020 08:59:08.927 [10548] DEBUG - [Transcoder] Throttle - Getting back to work.
When I check Titulia which runs on my Plex server it shows me I was connected using the local LAN IP.
So the server will duly apply the rules for remote playback.
Does your server VM use the host network? i.e. does it have an IP from your regular home network? You can see whether this is the case right around the beginning of the same log file.
You might also have an issue with ‘DNS rebinding protection’. I see that your server is “requiring” secure connections. So if a local client wants to access your server, it might be forced to use the remote interface because it has no other valid FQDN with a fitting security certificate.
See https://support.plex.tv/articles/206225077-how-to-use-secure-server-connections/
You can also try to set Secure Connections back to “preferred”.
There are also preferences in some clients (also in Android), titled “allow unsecure connections” which you can set to ‘in same network (as the server)’.
Restart both the server and then the clients after changing these preferences.
Re: VM the server and its hosts reside on the same private network.
DNS is fine as far as I can tell. I have run my Windows 2019 servers, DNS, Exchange, Web, etc. I use a wildcard SSL cert in my environment and didn’t want to get into with Plex on applying it.
And looking at the top of the log when the server starts up it looks like it talking to the server on 127.0.0.1 not the private IP of the server. Not sure if this is normal behavior or not.
Should I have Treat WAN IP as LAN Bandwidth enabled?
I see it mentions “Treat incoming requests from this network’s WAN IP address as LAN requests in terms of bandwidth. This often occurs when DNS rebinding protection is in place and clients on the LAN cannot contact the server directly but instead have to go through the WAN IP address.” I think turned it on without really reading the info. lol
And The Enable Relay
“The Relay allows connections to the server through a proxy relay when the server is not accessible otherwise. Note: this proxy relay is bandwidth limited” - I don’t have a proxy relay myself, does Plex have it’s on relay servers?
