I’ve added a bunch of TV Shows in a new library and I turned off video thumbnails for that library. When it starts getting indexed I noticed that it is looking at the entire media file as Plex is looking at the new library. I was hoping disabling the video thumbnails would avoid that.
I then when into the overall setting and turned off any type of video or other index that I could find. Rebooted the server. And it’s still reading the entire media file.
Is there a way to ensure that Plex doesn’t read the entire media file when indexing? Is it finding video pids or something? Can I disable that?
This video archive is on a network drive that is fast enough for streaming, but this index is gonna take forever. Would like to avoid downloading them all over the network.
When you edit the library in question… what directories are listed in the Add Folder tab?
Any overlap with the folders linked to other libraries?
When scanning a library, Plex will look for changes in all folders linked with the scanned library. Disabling video thumbnails will only stop your server from creating said video preview thumbnails for new items in that library – it’s not otherwise controlling what’s being scanned/analyzed.
It is copying the file over the network to the Plex machine. All of the files.
I disabled intro and credit detection. And voice activity.
I’m not sure what you mean by your last point. I think that to be able to get the format and analyze the container it should only need the first few hundred KB of the file. Is there something else that it does there?
Cool, I didn’t realize those settings were there. I’ve now disabled them. Though, I guess have to wait until tonight to see if it helps.
I think part of my issue is that I’m using the “Scan Library Files” or the “Refresh All Metadata” menu items on the library. Which may use a different set of settings? Is it possible to force a “find .” to run without downloading the files?
Generally speaking the scheduled tasks is what I want to use, but for testing to make sure things are working.
Ah, figured it out. Thank you everyone for all the information on Plex, I definitely had some things configured wrong there. But the root of my issues after getting the configuration right, were with davfs2 in Linux.
Apparently on file access it tries to cache the whole file. So even though Plex wasn’t requesting it it was downloading the whole file in the background.
What seems to be working now is using rclone with their mount command. Which does dav filesystems and you can configure how much caching is done, so it won’t download the whole file if it isn’t needed.
No, I implemented the filesystem support as DAV because I already had a webserver running and so adding it was pretty straightforward. Though NFS is probably a good idea in retrospect.
