Hi,
My Plex Media Server runs on a dedicated Arch server. Since one of the recent updates I can no longer resume playing a video via the web player on either OSX Safari, FireFox or WIN FireFox, Edge.
If I start playing a video close it and try to resume playing it later it will not resume, I can start playing that same video from the start without out any issues. If I pause a video and resume it will play for a while and stop, again I’m able to play the video from the beginning without any issues.
On iOS via the Plex app I can resume all videos with no issues, the very same video that can’t resume on OS X or WIN via the web player resume instantly.
Would you please recreate the issue, stop it, then capture the Log files please? Settings - Server - Help - Download Logs and attach that ZIP file here.
It looks like the transcoder, which is slowly transcoding ahead (in sloth mode) while paused, It hitting a really bad spot in the video file’s timestamps.
It gets to the point of asking for timestamps out of range for the whole file.
Aug 18, 2017 15:38:14.457 [0x7f19023fe700] DEBUG - Completed: [127.0.0.1:46100] 206 PUT /video/:/transcode/session/1ipx91s2vqxyjur9wlptpsou/1a8050e7-aff2-4322-9538-7bb50dccb1d9/progress?progress=39.6&size=-22&remaining=183&speed=2.9 (12 live) 7ms 326 bytes
Aug 18, 2017 15:38:14.458 [0x7f18f13fe700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 26091512 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.475 [0x7f18f5fff700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 26331112 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.489 [0x7f18f3ffd700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 26570744 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.519 [0x7f18f77ff700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 27051992 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.538 [0x7f18eafe6700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 27291640 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.563 [0x7f18f1bff700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 27772904 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.594 [0x7f18f0bfd700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 28254152 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.601 [0x7f18f13fe700] ERROR - [Transcoder] [mp4 @ 0x153e880] Application provided duration: -16 / timestamp: 28346360 is out of range for mov/mp4 format
Aug 18, 2017 15:38:14.603 [0x7f1902bff700] DEBUG - Completed: [192.168.1.5:51045] 200 GET /video/:/transcode/universal/dash/1ipx91s2vqxyjur9wlptpsou/0/14.m4s (15 live) TLS GZIP 1468ms 2027186 bytes (pipelined: 797)
I also see a file where the valid framerate 23.976 is being declared invalid because timestamps are incorrect here too.
Aug 18, 2017 16:20:32.775 [0x7f18e37ff700] DEBUG - MDE: E5 - The Package: Direct Play is disabled
Aug 18, 2017 16:20:32.775 [0x7f18e37ff700] DEBUG - MDE: E5 - The Package: media must be transcoded in order to use the dash protocol
Aug 18, 2017 16:20:32.775 [0x7f18e37ff700] DEBUG - MDE: E5 - The Package: no direct play video profile exists for http/mkv/h264
Aug 18, 2017 16:20:32.775 [0x7f18e37ff700] DEBUG - MDE: E5 - The Package: no direct play video profile exists for http/mkv/h264/aac
Aug 18, 2017 16:20:32.775 [0x7f18e37ff700] ERROR - get - invalid frameRate value: 23.976
Would it be possible to recreate the issue with only one playback instance happening? It would greatly simplify debugging
Rather than starting a new thread, at least for now, I’m having resume problems on every device
Nvidia Shield
Fire TV
iPad
Android phone
Media is stored on MacMini which is used as the PMS and the other devices connect either by wifi or ethernet. Never a problem before, very much so now, pretty much renders resume unusable and as such I’ve largely switched to using Kodi for local media which loads much faster and resumes without problem.
@Dudaai said:
Rather than starting a new thread, at least for now, I’m having resume problems on every device
Nvidia Shield
Fire TV
iPad
Android phone
Media is stored on MacMini which is used as the PMS and the other devices connect either by wifi or ethernet. Never a problem before, very much so now, pretty much renders resume unusable and as such I’ve largely switched to using Kodi for local media which loads much faster and resumes without problem.
Then I’m going to need your logs captured right after you recreate the issue. Settings - Server - Help - Download Logs. Attach the ZIP with your next post
@Dudaai said:
Rather than starting a new thread, at least for now, I’m having resume problems on every device
Nvidia Shield
Fire TV
iPad
Android phone
Media is stored on MacMini which is used as the PMS and the other devices connect either by wifi or ethernet. Never a problem before, very much so now, pretty much renders resume unusable and as such I’ve largely switched to using Kodi for local media which loads much faster and resumes without problem.
Then I’m going to need your logs captured right after you recreate the issue. Settings - Server - Help - Download Logs. Attach the ZIP with your next post
Hi. OK. Bit busy at the moment. Will do a log capture after I next get the problem. Will try with the known problem files later.
@Dudaai said:
Rather than starting a new thread, at least for now, I’m having resume problems on every device
Nvidia Shield
Fire TV
iPad
Android phone
Media is stored on MacMini which is used as the PMS and the other devices connect either by wifi or ethernet. Never a problem before, very much so now, pretty much renders resume unusable and as such I’ve largely switched to using Kodi for local media which loads much faster and resumes without problem.
Then I’m going to need your logs captured right after you recreate the issue. Settings - Server - Help - Download Logs. Attach the ZIP with your next post
OK. Just tried with the worst culprit. Movie called 13 Hours. At first it went fine. Then I realised I dialled down the video quality to 720 max to try and stop the problem. Switched it back to highest 1080 20Mbps (which should be OK given that PMS is running on a MacMini connected by Cat6 to the router, as is the Shield) and it just gave the spinning wheel. Closed Plex. Re-started it and and tried a resume again, I captured the log before the stream started which about 3-4 minutes.
Now I’m not sure if it’s something I’ve done wrong (with the conversion of the file) or if it’s something I’ve got set wrong or if it’s a problem caused with this Shield v6 firmware. I’d really like to get it fixed as it never happened before.
If this helps. I noticed (having never looked before) that there are LOADS of logs in the last zip. So for ease (maybe?) of understanding I moved all the old logs to another file and started fresh. Then I tried to play the same movie. Took around 3 minutes this time to ‘Resume’. Once it started I hit stop, waited a few minutes then downloaded the log file. Out of interest, should logging normally be enabled? If you’re not having a problem I guess not?
Further update. My PMS is 1.9.4.4325 running on a Mac mini running OS 10.13 and connected to router by CAT6. Shield is running Shield Experience 6.1 and is also connected to the router by CAT6 in the same room.
Once I start trying to stream this movie I notice this running in a loop: Failed to stream media, client probably disconnected: 32 - Broken pipe
Did some research on that and one thread suggested rebooting the modem, router and all devices on the network. Done that. Same. Smaller movies no problem. Also no problems with IMMEDIATE start of this movie on Kodi and IMMEDIATE resume from last played position.
Might have found something. Tried one more thing that I came across by mistake. My Shield is attached to a Samsung Soundbar with a separate subwoofer. As such I had Audio Passthrough disabled.
With the movie 13 Hours I went to settings from the movie info screen. Noticed the sound was set to AAC Stereo. So, I thought I’d try AC3 5.1. The app then advised Audio Passthrough needed to be on so I enabled it. Movie then started almost straight away from ‘resume’!! Tried the same with other movies that were buggy, same thing. Switch to AC3 5.1 and they start very fast, albeit not immediate.
@ChuckPA said:
While I cannot address your player issues, I can inform you PMS 1.9.6 was made available. I recommend upgrading to it.
Hi. Thanks? Very odd?? I never check to be honest. Just wait and at some stage while doing things on the server and it’ll say ‘Update Available’. It recently did to update me to 1.9.4.4325. Even forcing a ‘Check for update’ it says ‘Up to date’. Yet, as you say, I go to the download page and it’s 1.9.6.4385? Maybe it’d go so far behind in releases it can only do one at a time, not jump straight to the latest. Very strange. Anyways, updated now.
@Dudaai said:
If this helps. I noticed (having never looked before) that there are LOADS of logs in the last zip. So for ease (maybe?) of understanding I moved all the old logs to another file and started fresh. Then I tried to play the same movie. Took around 3 minutes this time to ‘Resume’. Once it started I hit stop, waited a few minutes then downloaded the log file. Out of interest, should logging normally be enabled? If you’re not having a problem I guess not?
So, is there a plan to look at these logs or what???
