Rclone+Unionfs+PlexACD resume playback problem

Hi everyone,

Im using rclone mount, with unionfs and Plex. Everything is going perfect, but when I have to stop playback, and continue later, it always give a timeout.

I’ve been doing debugs, and I saw the rclone always starts the playback from the begining, even when I want to go to minute 65. If I stop the playback again, and I want to start from 65 minute, rclone starts the transfer again.

Is there any way to do caching or something like that? It’s almost impossible to resume playbacks in movies.

Thanks in advance

I’ve been unable to stop a video and restart it from any point (even 15 seconds into a video) when using Plex and ACD. I’ve been experimenting with GoogleDrive and that single change got seek, restart, and very short start times all working.

I tested with another movies, and I can resume playback in some movies. Every movie has the same resolution, I don’t know what is happening.

If that’s the case, I’d check and see if there’s a consistency between transcoding/directPlay/directStream and if/when you’re able to resume.

In other words, does it always work when Plex is transcoding? Never work? Always work if it’s directPlay? etc.

I tested movies with the same resolution in original quality, one direct stream and one transcoding, and only works with direct stream.

Got the same issue; has someone better experience in a AWS setup?

I’m experiencing the same problems:

  • Resuming a movie will often fail entirely.
  • After the movie starts, the stream will often stop loading after a few seconds and crash.
  • When seeking in the movie, the movie will sometimes just continue and ignore the seek.

Streams usually work flawlessly when starting from the beginning and not stopping, albeit the --buffer-size 1000M buffer will run dry at times.

I’ve not been able to get some decent logs on this issue. If anyone has, I’d be interested to see what is happening behind the scenes!

I dont have those problems, I only have the resume playback problem, when transcoding. If the playback is direct play or direct stream everything goes ok, but it takes about 30 seconds to resume.

I think the solution is to cache content in local, I mean, it would be perfect to have new flags like, --cache-local 10G --cleanup-cache 3600s

With those flags the movie or whatever will cache 10G during one hour, and If the playback fails the user could resume movie without downloading it again from ACD.

1 Like