I run Plex in a docker container on Unraid 6.6.6. I use the docker from Plex inc. (I assume that’s the one from Plex). My Plex server streams to all other devices on network (roku, chromecast, computer, PS4) and it streams outside my network to a few friends all over my state.
What is a NW config/routing issue?
This basically means that the PMS doesn’t know how to get to 192.168.0.163 (whatever device that is). What IP does that map back to?
I haven’t worked with Unraid+Docker setups personally, but I found this article that seems to have some good information on how to setup the configuration.
Checked my router through the my desktop interface and that IP doesn’t exist on my network. My router has a mobile app as well and it shows that IP address as the Nvidia Shield but I made the shield’s IP static as 192.168.0.12. Could that be the PMS running on the Nvidia Shield getting it’s own address?
Attached are three additional logs from this morning.
MakeMKV - Blu-ray rips:
iTunes Rip - Futurama:
iTunes Rip - SG-U:
(Not in this log) I ran handbrake through the Futurama episode and it played fine on the Shield TV using Super HQ settings, although I had no audio, but maybe that’s user error.
Plex Media Server.log.zip (349.5 KB)
Apple TV Log.zip (580.2 KB)
Shield TV Log.zip (231.1 KB)
So a few things to keep in mind re: 2 ip’s.
If you had your shield connecting via wifi, and then changed to hard wired, you will get 2 different IP’s from that alone. Each one has a different MAC address, which means different IP.
If you manually changed to a static IP via the Shields NW interface, then your router and other devices connected to it could have had the old IP address in their DNS Cache. This will eventually age out of the caches.
You are generally better off setting a DHCP reservation (preferably outside of your DHCP scope range) on your router for your Shield. That way it will still get the “static” IP, but makes sure the router has more awareness from a local DNS,etc. perspective.
The NVShield PMS Server IP is based on the host IP, so that won’t matter at all.
Based on your testing results re: Futurama, that definitely points to something in the way that the iTunes rips are being ripped that the Shield isn’t liking.
One way to test if it is Plex specific would be to copy both versions of the Futurama files to a local thumb drive/USB hard drive and direct connect that the NV Shield and try playing them via a native ShieldTV app.
If you still see the same behavior, then the issue is like an issue related to the Shield TV’s Exoplayer utility.
I have VLC on the Shield would that work, or do I need to use a different app?
I did look at the info as it played, it was DirectPlay w/ Exoplayer v2 (i believe there was a v2 at the end).
I would have to double check if VLC on the Shield is using Exoplayer v2 or it’s own engine. It would definitely be a good starting point test though.
I believe ExoPlayer v2 came out when Shield OS 7.x was released, but not 100% positive.
With TuneKit M4V Converter, I can rip;
Is there a preference?
I would probably use MP4 HD or MKV HD. But those are just containers, you really need to know the underly codec’s being used.
You definitely want Video Codec to h264, what are your audio codec options?
Auto or AAC
Typically AAC is fine.
I’m only connected via wifi but i think it might have had something cached for the 192.168.0.163 because that’s what the shield defaulted to when I turned it back to DHCP. I then set the static in the DHCP section of my router. That seemed to clear up that error but it continues to stall while playing. I downloaded these logs just after it crashed. I have one other person streaming from it at the same time. His is transcoding and his continues to play while mine froze. I was watching the console to see if an error popped up but it didn’t, it just continued chugging along like it was still playing. Is there anything I should be looking for in the logs when this occurs?
Glad it sounds like we have the 163 IP issue sorted out.
In your latest logs, I see several types of different errors.
There is one related to a what I assume is an add-on channel called MyPlex.
10060 Feb 12, 2019 12:55:11.903 [0x1539ecbff700] WARN - MyPlex: Invalid connection URL 'plex.roguemedia.us'.
10072 Feb 12, 2019 12:55:14.412 [0x1539ea1c9700] WARN - MyPlex: Invalid connection URL 'plex.roguemedia.us'.
There are also some error related to performance on the Docker container/PMS.
13440 Feb 12, 2019 13:11:03.942 [0x1539cfffb700] DEBUG - Request: [::ffff:192.168.0.119:51964 (Subnet)] GET /hubs/home/recentlyAdded?type=8 (25 live) TLS Page 0-15 GZIP Signed-in Token (xhonestly)
13441 Feb 12, 2019 13:11:03.943 [0x1539cfffb700] DEBUG - Setting container serialization range to [0, 15] (total=-1)
13442 Feb 12, 2019 13:11:03.943 [0x1539daffd700] DEBUG - It took 50.000000 ms to retrieve 200 items.
13443 Feb 12, 2019 13:11:03.944 [0x1539ec1fe700] DEBUG - Completed: [::ffff:192.168.0.119:51964] 200 GET /hubs/home/recentlyAdded?type=8 (25 live) TLS GZIP Page 0-15 2ms 553 bytes (pipelined: 2)
13444 Feb 12, 2019 13:11:03.949 [0x1539e47ff700] DEBUG - Jobs: '/usr/lib/plexmediaserver/Plex Transcoder' exit code for process 24092 is -9 (signal: Killed)
13445 Feb 12, 2019 13:11:03.970 [0x1539daffd700] WARN - SLOW QUERY: It took 107340.000000 ms to retrieve 50 items.
13446 Feb 12, 2019 13:11:03.983 [0x1539ec1fe700] DEBUG - Auth: authenticated user 1 as xhonestly
I would make sure you you are running the scheduled maintenance tasks:
https://support.plex.tv/articles/202197488-scheduled-server-maintenance/
Especially the DB optimize, clean bundles, etc.
Also reality is that if you have the Library DB living within the Docker Container, the overlay filesystem uses incurs some performance penalty. This is really going to depend on how you configured your Plex docker container when you set it up.
See here for an example of what I mean.
https://hub.docker.com/r/linuxserver/plex
In the example command line, they show 4 different volume mappings for Plex (1 for Library, 1 for Transcoding, 2 for your Media)
How is your container setup?
I’m using the Plexinc docker. It has data, transcode and config (this is the directory for the PMS). The myplex is a reverse proxy through let’s encrypt to access the server from outside the network. The reverse proxy is for the other containers I’m running and I think my brother just set up the plex one because he could. I just switched my docker from the unraid version to Plex inc. because I read a few blog posts that the unraid one was being discontinued so that may be why there are performance issues. I have the scheduled maintenance tasks set to run I just don’t think they have for this one yet.
But again, the only issue I am having is with the Shield version of plex. I just watched a video on my web ui and it was fine. I watched a video on my android phone the other day, fine. I pretty much watch 4 hours of TV a night on my Roku Plex app all without issue. But if I play anything from SD to UHD on the shield it freezes after a while. I just tried forcing transcode and it continues to play without issue. Looks like it must be something having to do with the Shield itself and not my server. There are tons of people having the same issue I am. It’s been ongoing since November, ever since the latest update. There hasn’t been any further updates from Nvidia or Plex on this issue.
So if the re-ripping fixes the Shield issue, does this mean its a Exoplayer v2 issue and not a “slow network” issue as Plex warns? Since Apple TV has no issue with any file…
Trying to keep the issues straight here 
If the re-rip fixes the issue, it either comes down to either an ExoPlayer 2 issue or the first rip being out of spec for what ExoPlayer v2 expects. Some may be more forgiving with not exactly following the letter of the law vs. others.
Then we can do a quick dump of the media properties to see where the differences are.
You can use MediaInfo or via Plex this way : Investigate Media Information and Formats | Plex Support
I really need the clients logs from the Shield TV along with the Server logs when this is occurring.
See here on how to dump them: https://support.plex.tv/articles/201573117-android-android-tv-fire-tv-logs/
Also Plex just released a new shield beta client this week or late last week.
See here for how to get access to it.
https://support.plex.tv/articles/201844613-early-access-beta-releases/
Based on a new Nvidia post I just found, it looks like the issue may being caused by an issue with the video files encoding themselves. https://forums.geforce.com/default/topic/1093822/shield-tv/shield-experience-upgrade-7-2-2/post/5973464/#5973464.
I am checking with some of my Plex counter-parts on this to see if they agree or not.
