I think I’m seeing the same. Thanks so much for all the useful testing you have done. Curl inside the docker works but logs full of errors. Will try adding to docker resolver and rollback then report.
My first rollback fixed it. binhex/arch-plexpass:1.22.3.4392-1-01 transcodes finally. That’s 10 hours down the drain…
Local resolver didn’t change anything.
Please detail for me what you did? (presume I’m clueless)
I want to recreate this step-by-step for engineering to see. (They want to see it)
Reviewing the “GoodPlex” and “BadPlex”, I see one thing immediately.
May 08, 2021 11:57:37.945 [0x7f3ea0f6db38] DEBUG - Starting plug-in /usr/lib/plexmediaserver/Resources/Plug-ins-a1b1f3c10/PlexMusic.bundle.
May 08, 2021 11:57:37.945 [0x7f3ea0f4ab38] DEBUG - Starting plug-in /usr/lib/plexmediaserver/Resources/Plug-ins-a1b1f3c10/Framework.bundle.
May 08, 2021 11:57:37.945 [0x7f3ea0f27b38] DEBUG - Starting plug-in /usr/lib/plexmediaserver/Resources/Plug-ins-a1b1f3c10/WebClient.bundle.
May 08, 2021 11:57:37.946 [0x7f3ea0fc0b38] DEBUG - Waiting for 24 plug-ins to finish starting.
May 08, 2021 11:57:37.947 [0x7f3ea0f6db38] DEBUG - Starting file watcher for tv.plex.agents.music
May 08, 2021 11:57:37.947 [0x7f3ea0f6db38] DEBUG - Plugin: setting environment variable: 'PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python'
May 08, 2021 11:57:37.947 [0x7f3ea0f6db38] DEBUG - Plugin: setting environment variable: 'PYTHONPATH=/usr/lib/plexmediaserver/Resources/Python/python27.zip:/usr/lib/plexmediaserver/Resources/Python/lib/python2.7/site-packages'
May 08, 2021 11:57:37.947 [0x7f3ea0f27b38] DEBUG - Starting plug-in /usr/lib/plexmediaserver/Resources/Plug-ins-a1b1f3c10/TheMovieDB.bundle.
May 08, 2021 11:57:37.947 [0x7f3ea0f4ab38] DEBUG - Starting plug-in /usr/lib/plexmediaserver/Resources/Plug-ins-a1b1f3c10/LocalMedia.bundle.
May 08, 2021 11:57:37.947 [0x7f3ea0f6db38] DEBUG - Spawned plug-in tv.plex.agents.music with PID 495
May 08, 2021 11:57:37.947 [0x7f3ea0f6db38] DEBUG - [tv.plex.agents.music] Sending command: GET /:/prefixes
May 08, 2021 11:57:37.947 [0x7f3ea0f27b38] DEBUG - Starting file watcher for com.plexapp.agents.themoviedb
May 08, 2021 11:57:37.947 [0x7f3ea0f27b38] DEBUG - Plugin: setting environment variable: 'PYTHONHOME=/usr/lib/plexmedia
Who’s setting this variable?
May 08, 2021 11:57:37.947 [0x7f3ea0f6db38] DEBUG - Plugin: setting environment variable: 'PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python'
It does not exist in my Docker container or the “GoodPlex” shown here.
If it was me you are asking, no idea. It says it’s a plugin. The plugins I have installed are Sub-Zero and Trakt.
Have just now uninstalled them for a further test. Will update when done.
Also I noticed I get the following line in my Docker logs for the new version only (no idea if it’s relevant):
Critical: libusb_init failed
Uninstalling the plugins didn’t help at all. I can still actually watch something off it’s via direct play - anything that requires a transcode fails as Plex can’t retrieve the XML file the codec from Plex.tv.
In terms of what I did:
Stopped/removed my working Plex container. All data is saved to the host.
Update the docker-compose.yml to use the plexpass tagged image, instead of the public tagged image.
Launch the new container. That’s it.
Looking through the new logs, that variable gets set by various agents:
Plex Media Server.log:May 10, 2021 00:29:58.402 [0x7fe1b27d2b38] DEBUG - Starting file watcher for com.plexapp.agents.themoviedb
Plex Media Server.log:May 10, 2021 00:29:58.402 [0x7fe1b27d2b38] DEBUG - Plugin: setting environment variable: ‘PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python’
Plex Media Server.log:May 10, 2021 00:29:59.567 [0x7fe1b27d2b38] DEBUG - Starting file watcher for com.plexapp.agents.opensubtitles
Plex Media Server.log:May 10, 2021 00:29:59.567 [0x7fe1b27d2b38] DEBUG - Plugin: setting environment variable: ‘PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python’
Plex Media Server.log:May 10, 2021 00:30:01.868 [0x7fe1b27d2b38] DEBUG - [com.plexapp.agents.fanarttv] Killing.
Plex Media Server.log:May 10, 2021 00:30:01.868 [0x7fe1b27d2b38] DEBUG - Plugin: setting environment variable: ‘PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python’
Plex Media Server.log:May 10, 2021 00:30:02.683 [0x7fe1b27d2b38] DEBUG - [com.plexapp.agents.opensubtitles] Killing.
Plex Media Server.log:May 10, 2021 00:30:02.684 [0x7fe1b27d2b38] DEBUG - Plugin: setting environment variable: ‘PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python’
Plex Media Server.log:May 10, 2021 00:30:03.953 [0x7fe1b27d2b38] DEBUG - [com.plexapp.agents.plexthememusic] Killing.
Plex Media Server.log:May 10, 2021 00:30:03.955 [0x7fe1b27d2b38] DEBUG - Plugin: setting environment variable: ‘PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python’
grep: PMS Plugin Logs: Is a directory
Plex Media Server.log:May 10, 2021 00:30:06.232 [0x7fe1b27d2b38] DEBUG - [com.plexapp.agents.htbackdrops] Killing.
Plex Media Server.log:May 10, 2021 00:30:06.233 [0x7fe1b27d2b38] DEBUG - Plugin: setting environment variable: ‘PYTHONHOME=/usr/lib/plexmediaserver/Resources/Python’
Glad I found this thread, at least it means I didn’t bork something up… I had just done a Firmware update to my server at the same time I updated Plex on the QNAP…
I’ve been trying to figure out what the heck I did worng…
So… same issue here using plex on qnap.
rolling back 1.22.0.4163 until this is sorted.
I’d be in over my head trying to keep upw ith what y’all are working on here…
With the added DNS nameservers, most things seemed to work again. However now I’m also often getting full Plex crashes/restarts when starting to stream media. On top of that I have no Playback through my browser. On Xbox, Chromecast and via Media Player on Windows still playing, but the crashing is getting too regular.
Can someone with this problem copy the host OS and docker container version of /etc/resolv.conf and upload it here?
Host OS /etc/resolv.conf:
nameserver 10.0.0.1
Docker containter /etc/resolv.conf:
nameserver 127.0.0.11
options ndots:0
The new public version 1.22.3.4523-d0ce30438 also works without issue.
Any updates on this one? Or requests for further testing/logs?
Just tried again, and the latest 1.23.0 build still fails to be able to resolve plex.tv whereas any 1.22 build can.
Direct play works but any attempt to transcode fails as codec details can’t be relieved retrieved from Plex.tv.
I’ve tried the latest 1.23.1.4528-c0513eb4c build, and have the same curl issues; works on 1.22.x, but not 1.23 versions.
To note, I’m not using docker, just straight RHEL 7.9… no IPv6
Yes, I still have the same issues with the new build - 1.23.1.4528-c0513eb4c.
Still the same with new version: 1.23.2.4584-bdefd2026
And now that Public has updated to 1.23.1.4571-6119e8eed, this doesn’t work for me either.
So basically Plex is now completely broken for me.
Was hoping by reporting this issue during “beta” there would be some traction to fix it before it hit public.
Edit: In the container logs, for the failing versions I also get the following line a little while after start up:
Critical: libusb_init failed
Is this 100% reproducible for you and, if so, might we be able to arrange for some Teamviewer or SSH time on the machine to deep dive ?
This DNS issue will still exist until we find it. The problem has been getting access to a machine which displays the problem.
Yes, completely 100% reproducible.
Its just really bizarre. 1.22 works fine 100%. 1.23 will work for direct play only, anything else fails due to the DNS issue. Same Docker network etc, same bind mounts etc.
Might we be able to get some time with you on that machine?
Sure. The machine where PMS is running (under Docker) is headless, but I can share a screen from my desktop showing me SSH into the box, changing between working and non-working images and what happens in logs and when trying to access Plex.
I’m going to reach out to our build team expert on this.
She is on the build team and will be the one performing the fix.
After chatting with her, I will reach out to you in PM and we can continue from there.
3 Likes
To all reading here:
Working with Ridley, we’d like to conduct a test on those systems which cannot resolve plex.tv when using plex.tv
Specifically:
After some investigating, we’re onto something.
Please do the following:
dig plex.tv AAAA
and report the result. You should get
- NOERROR response with an SOA record
- a SERVFAIL message
This is a good reply.
[chuck@lizum ~.502]$ dig plex.tv AAAA
; <<>> DiG 9.11.31-RedHat-9.11.31-1.fc33 <<>> plex.tv AAAA
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 21718
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1
;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 65494
;; QUESTION SECTION:
;plex.tv. IN AAAA
;; Query time: 0 msec
;; SERVER: 127.0.0.53#53(127.0.0.53)
;; WHEN: Wed May 26 21:30:43 EDT 2021
;; MSG SIZE rcvd: 36
[chuck@lizum ~.503]$
Next, Please tell us your modem/router vendor.
We think we’re seeing a commonality and would like to gather more information before digging into it futher.
1 Like
