Recommended Dropbox (Formally Google Drive) and Plex Mount Settings

@Animosity022, on your github page, when taking about your mergerfs setup, you mention that you use cache.files=auto-full as it replaced auto_cache. In the bit of code above the explanation, though, you have cache.files=partial listed. I don’t know enough about the setting to know the difference, and I’m wondering which one is actually preferred. Thanks.

The mergerfs settings are documented here:

You can pick based on what you feel is needed for your setup.

Can I ask, why did you stop using?

--drive-chunk-size xxxM \
--vfs-read-chunk-size xxxM \
--vfs-read-chunk-size-limit off \
1 Like

For the first, I don’t upload directly to my mount so there is no need.

For the second two, the values I use became the defaults so I just use the defaults now.

1 Like

this guy is implying that the jellyfin bug is your mergerfs settings. might it be possible? https://github.com/jellyfin/jellyfin/issues/2276#issuecomment-574004082

Jellyfin is pretty much the same thing as Emby (at a high level as it was what it was based on). Which both do the same things as Plex.

I can't use direct_io as I use torrents so that would break torrenting on the mount.

Not sure what it meant by non_empty as that's the same reason for me for not over mounting things.

It looks like they have an old version as they have many old options set that are replaced and a ton of defaults that aren't needed. Would doubt it is my mergerfs command or rclone command but you'd rule that out by checking the logs for each and seeing what's going on when the issue happens.

Is it possible to see the contents of this with rclone?

Yes, with something like:

rclone ls GD:  --drive-trashed-only

You can even mount with:

felix@gemini:~$ rclone mount GD: /Test --drive-trashed-only
1 Like

I'd suggest to make a new post as this has nothing to do with my settings as you aren't using them. Use the question template and ask there rather than hijacking my thread. Thanks!

any special reason jelyfin/emby seek through videos on direct stream faster than plex does?

Not sure I'm following the question.

If I'm direct streaming, it's either playing or not. I have no issues direct playing from Plex or Emby. Both work with no issues for me.

My use case tends to be watching TV or a Movie so I rarely seek. I've done some ad hoc testing and found just about no different between Plex or Emby.

Emby/Jellyfin have no HDR support on ATV for direct play so that's a deal breaker for me.

If you have a separate question about something other than my settings, you can always make a new post and other folks can chime in on your question and provide their feedback as well.

well direct olay doesnt work on browser which is what i use but that isn't what i'm asking/talking about

my question is more about the fact that when i skip/seek through video on plex(direct stream) its slow but when i seek/skip through video on emby/jellyfin(direct stream) it's fast.

wondering why that is.

Your best bet would be to open a new post, share your settings, versions and a few debug logs from each and I'm sure we can shed some light on it.

How often do you play content from Plex in web browsers?

Never.

I use ATVs in my house, iPhones and the native Plex App if I had to slum it and watch on a computer.

Hi @Animosity022,

I'm using some of your scripts in my build but I face an issue when radarr is getting a newer version of a movie: once the grabber has complete the download, radarr cannot access the first file to delete it I assume: I got an access denied in radarr log when importing the movie.

**Couldn't import movie /downloads/xxxxxxx.mkv: Access to the path "/movies/xxxxx (2016)/xxxxx.mkv" is denied.**

My radarr is running in a docker with the volume /movies/ (in the container) mapped to my host folder /gmedia/. This folder is mounted with mergerfs and has the following structure:
/gmedia/
/temp/ => local path on my host where all media freshly grabbed are stored
/gmount/ => gdrive mount (crypt & cache)

when I do ls -l to check ownership of files in docker and on my host, all files are owned by root -rw-rw-r--

Do you have any idea what more can I check ?
Thank you

Sorry @Plouf as I don't use dockers at all as I find them unneeded as I run all on a single machine. They only add unneeded complexity for me.

Sounds like you are having a permissions issue related to the docker setup.

understand
There is one thing i want to confirm: with your scripts, radarr can delete a previous version of a media and replace it by a better one ? let's say it deletes a 1080p and replace it by a 4k version ?

Yep. I do that all the time.

My rclone mount is read/write so it deletes from the rclone and it's replaced on the local disk. Works like a champ as that's one of the main use case setups I was trying to solve.

ok, I will investigate.
I wonder if I should change umask 002 in rclone mount command to 022