Option to sync ALL metadata and libraries to iCloud

Infuse will currently sync corrections made in a library to iCloud, but everything else remains on each device. This means that every device is required to re-fetch all the metadata and library matching. This approach is fine for small libraries, but it’s inefficient, time-consuming, and borderline impossible for large libraries.

In order for my Apple TV to index everything, it took two days of leaving it on and not watching anything, because Infuse stops fetching while streaming. In that time, I was also API banned from Google Drive for 16 hours, as my entire library is located on that service; it should be noted that when API banned, Google Drive does not let you stream anything out of it, either, so I wasn’t allowed to watch anything during that time. Obviously, this is a huge deterrent in reproducing my libraries elsewhere. I simply won’t do it, because I don’t want to get API banned again, which I absolutely will. Multiply this by all my devices, and it’s clear this is not the best approach. I also just don’t want to have to keep every device open and effectively rendering them unusable during that indexing period.

I really want the ability to configure Infuse to upload the entirety of my Infuse library index and metadata to iCloud. I get the reason for only syncing corrections to iCloud right now, but there really should be an option to allow everything to sync up to iCloud for those of us who don’t want to waste hours or days waiting to reproduce our large collections. Personally, a few GBs in iCloud is nothing. I would gladly pay the $1.29/mo to Apple’s iCloud subscription (if I had no space left) for 50GB of storage if it meant that all my Infuse libraries, indexes, and metadata were saved. Provide a warning in the settings, so people can’t accidentally enable it without knowing what they’re doing, and voilà. This would mean that all of the work of getting Infuse to a usable state could be reproduced on multiple devices with very little effort, and, most importantly, very little time. I don’t see any other way for Infuse to be able to scale with larger libraries.

I say everything should be synced. This means that when iCloud downloads the dump to a new device, that device will show all the libraries and all their metadata–without need to scan the underlying directory structure. If a file has been deleted since the dump, then it can be marked as unavailable. Plex handles it similarly, and the only way to remove dead files is to hit the “Empty Trash” option. We can then choose to run a new scan, but it shouldn’t auto-delete these files, and only add new ones.


(I have 1 TB iCloud storage, ATV3, multiple iPads and iPhones)

1 Like

Absolutely agree. I felt the pain of trying to set up my Apple TV / iPad / iPhone. 1577 movies and 13k tv episodes has been extremely painful.


That would be nice. I have 200GB in iCloud, or alternatively a lot more in Gdrive (already connected to Infuse). If Infuse could set a folder there for the artwork/metadata and the paths for Google Drive that would help alleviate a lot of the issues using Google Drive.

How is downloading a large or small set of meta-data from iCloud for each device faster then downloading same meta-data through the through current mechanism? You still have to download something for each device.

You don’t see the obvious benefit? Downloading is not the issue. Scanning is the issue. If all the hard work of scanning, which takes days for my (and other people’s) libraries, could be reduced to a single download of several GBs that will take a couple minutes at most to download, the benefit is clear.

In your opinion, how is this not faster? You think it’s faster to scan tens of tens of thousands files in a filesystem (or cloud service), then makes tens of thousands of requests to an API to associate the metadata to them, and to do this on every device? It’s infinitely faster to download the resulting payload of all that scanning and metadata associations work, which is a single, dumb operation, versus all of that work.


It still as to scan all the files on your share regardless to ensure they didnt change, then re-sync/download meta-data. Not sure about you, but ppl delete/add files constantly and Infuse does not display meta-data for files that dont exist.

If you dont like the scanning per device, there’s this client/server called Plex.

You seem to misunderstand the purpose of the suggestions forum. A post made in it implies that I disagree with how something behaves or performs, and I therefore want it to behave or perform differently–preferably as described in my suggestion. My suggestion is not a misunderstanding of how it currently behaves or performs, which you appear to understand well enough. If it were to continue to behave as you describe, then that means the suggestion hasn’t been implemented, which means it’s still just a suggestion. If it gets implemented, then it’s a new feature. You see the point I’ve made? You’re wasting everyone’s time with your pointless criticism, based on an unfortunate misunderstanding of the point of this thread. Try not to drag it out any further.


“Scanning is the issue. If all the hard work of scanning, which takes days for my (and other people’s) libraries, could be reduced to a single download of several GBs that will take a couple minutes at most to download, the benefit is clear.”

Plex. Nuff said.

You should start a new thread in the suggestions forum–and suggest that the Firecore team cease development of Infuse because Plex. I’m dumbstruck that I hadn’t thought of it first. You beat me to it. You’re right, though. There can’t be two pieces of software that do what I want. Two is absurd.

1 Like

This would be really great. +1

I just added a new device and I’ve been banned (temporarily) for hitting download quota (https://i.imgur.com/QY6xL3v.png), as expected. You can determine this a couple of ways (How to know if 24-hour banned from Google Drive? - Help and Support - rclone forum), but the easiest is to just log into the Web interface for Drive and attempt to download a large file. Once you’re familiar with the issue, though, you’ll know you’re banned when you just can’t stream anything to any of your devices anymore; you’ll be greeted by various client errors. The ban typically lasts between 12 and 24 hours, but many have reported longer waits. Infuse filesystem scanning is too intense and sub-optimal. This needs to be improved. When Plex Cloud was first released they also had these problems, but eventually resolved them, as did plexdrive, a third party project. In a different ticket to improve Google Drive scanning (Improve performance of Google Drive scanning), it’s mentioned that the solution to the problem is to look at the Drive change list. I’d be very surprised if the Infuse team told me they’re using Google Drive’s API to perform scans, because the performance is really on par with basic recursive directory traversal techniques, which is really not optimal when it comes to scanning cloud services, which are not filesystems: they just emulate directory structures. Based on the outward performance, it behaves like the Infuse team wrote the codebase for scanning directories, and later on, once it was figured out how to connect to Google Drive, that same codebase was pointed to the Drive. While that is okay to get things going and saves time, that will definitely not scale, and a new scanning codebase will need to be written specifically for Google Drive, and any other cloud service provider offered. Given that Google Drive has by far the most use (because it has by far the best pricing), this treatment should be prioritized for it first.


I really hope someone at Firecore/Infuse is paying attention to your thread because we really need this.


I second this request. I have multiple Apple TVs. It takes several minutes when I launch Infuse for the app to scan my library, find new files, and then assign the metadata from the cloud backend. There should be a background sync happening between TV 1 to iCloud and then iCloud to TV2 so that new files are waiting for me in my library.


1 Like

I’m extremely jealous that it only takes you several minutes. Regardless, I think we can both agree that’s several minutes too long!


In general I still support this topic but I’ve also Plex on my server and this works only GREAT!!

+1 for Metadata in iCloud, with an SQL database. That wouldn’t be a big size and it would contains watched infos as well.

Other hand about iCloud sync: I have an ATV and some iPhone in my family. I don’t use iCloud because on ATV use NFS connection, on iPhones use WebDav. If I turn on the iCloud in Infuse set it both connections to all devices, than everything appears double.
Unfortunatelly currently iCloud useless for me.

Yeh, totally agree. This feature would be extremely beneficial.

I’m also wondering why there has been no response from the Infuse Team on this matter? If there is a valid reason why this can’t or won’t be implemented, than it would be good to be made aware of this from an official source.


The roadmap thread (at the top of this sub-form) does show that “improved iCloud sync” is a planned feature upgrade for the 5.8 release although it gives no details about what this includes.