Bandwidth whilst adding content to library

I’m thinking more about hosting items off the lan, assuming decent speed connections, and wondered if anyone understood the process of osmc/kodi adding the files to the library.

I’m assuming it’s dealing with metadata so bandwidth should be fairly low

Also whether virtual file systems plugins for rar files makes any difference there
This seems to say it doesn’t need to unpack the files - https://kodi.wiki/view/Add-on:RAR_archive_support

I appreciate there are choices here but options I’m considering in a rough order of preference:
Mounting files system through sshfs
Wireguard
kodi on fileserver for ‘local’ scraping to shared sql server
backend - jellyfin or similar thought rather avoid this if possible

Anyting else I’ve missed

Thoughts most welcome please and thanks for reading

I’m not sure you made it particularly clear what your trying to accomplish and what hardware your working with. Your talking about LAN access but then you bring up Wireguard. If you were to clarify what your planning on hosting your files from and what client endpoints you are wanting to use with it then it may allow for appropriate suggestions to be made.

Sorry, didn’t think the hardware was that important as was just trying to understand the process/ requirements for adding media and the bandwidth involved. I should probably just test or see if I can understand the code, the latter is less likely

“Hosting items off the lan” = remote WAN fileserver.

Hardware:
Mainly Pis with Osmc but may as well assume kodi on laptops as well.
Fileserver- whatever it needs to be but linux of some flavour

Reason:
Family has osmc across several households I manage and means I don’t have to sync files to local networks

Benefits
Reduced data management requirement
One version of the truth

Negatives: Possible Bandwidth - hence the question.
I don’t have a problem with streaming the media here, but wanted to avoid large data charges if the actually adding to the library uses large amounts, that’ll change whether I let each device scrape the media or centralise/ sync the databases

I think you will find that Jellyfin, Plex, or similar software is the path of least resistance. As far as bandwidth goes the metadata wouldn’t be expected to be enough to bother factoring in. The issue is normally with the media itself unless you happen to be fortunate enough to have a large upload bandwidth from your ISP (most peoples internet connection at their home is much slower on the upload than download). The bespoke applications for doing this type of sharing not only have the advantage of easier deployment, but they normally will transcode on the fly to fit a limited network connection.

This solution doesn’t have to be all encompassing. I run Kodi across my house using a Win10 PC for a file server and MySQL with all the clients connecting to that. It also runs a Plex server that my daughter connects to over the internet. The two systems are not tied together other than them both scraping the same files. Plex auto scrapes so it just sits there in the background and I rarely have to touch it or think about it.

Thanks, so I’m hearing it’s kodi is only ever using metadata until it comes to actually playing. Great news.

One exception to that I can think off, and writing here as a note to self, is kodi unpacks compressed rars locally so I’d need to deal with them differently.

On the solution, thanks for suggestions, jellyfin is tempting just for the http playing (looking at you chorus) and I suspect it could tie in with some android audiobook playing but I’ve a natural preference to keep as things as low level as possible, filesystem, vpn etc.

So I need to ponder the options but that’s been really helpful and saved a lot of time, thank you!

The details are going to depend on exactly how you have it setup but if your talking MySQL the thumbs are stored local and loaded on demand when not cached. Where this bandwidth comes from depends on if you scraped local or online art. By this I mean that in a typical setup you are just scraping from an online source. In this case Kodi stores the url of the the scrapers web site. When a client goes to load missing art it grabs this url and then downloads from online so this bandwidth wouldn’t be coming from the file host. If you had stored local art that you scraped then it would.

As for bandwidth from scraping this shouldn’t be a thing you would want to be doing remotely if your doing MySQL and shouldn’t be needed. You scrape on a local client and remote clients can just use that. If you needed to tweak file paths for locations you can just do that with path substitution instead of trying to have separate libraries.

No clue about the RAR’s. I’ve never used them nor could understand why anyone would want to keep them like that. To each their own.

I would expect issues if your going to try to MySQL over a WAN though. Kodi gets a bit bitchy when the latency goes up. If your lucky the only major issue will be the watched status not updating all the time in the UI, but I suspect you will get crashes as well.

I think if I wanted Kodi to look like Kodi and a remote client connected like this I would probably just use Plex all the way round and then just run PlexKodiConnect to hide the backend.

As darwindesign suggests the easiest way is probably using Plex etc.

It is possible to run Kodi over WAN. I’ve run Kodi over SSH tunnels (using sshfs) and it works reasonably well. Some tips if you chose this route:

  1. The upload bandwidth on the internet connection where you are storing media should be at least 50% higher than average video bitrate. Kodi is not very forgiving if it doesn’t have sufficient bandwidth.

  2. Keep to local sqlite database. Don’t run shared MySQL. Kodi interface performance is tied to SQL lookups. The latency on remote connections is enough to noticeably slow it down.

  3. Best to use .nfo files for metadata. Default settings in Kodi is to load video to extract thumbnails which will be slow. The settings you want to disable are:

In Settings > Media Settings > Videos

  • Use video tags = None
  • Extract video information from files = No
  • Extract thumbnails from video files = No
  • Extract chapter thumbnails = None
  1. On the remote devices in the sshfs setup you’ll want to use keepalive messages for the service to detect disconnections and to reconnect. Using sshfs and AutoSSH can help keep the tunnels ssh up.