Why don’t you put the library on a local server? Other system like Zidoo are doing it like this, and it works wonderfully. A complete restore of the library, in my case around 1500 movies and series, takes less than 30 secs.
Several of us, including me, LOUDLY, have been BEGGING for this feature for YEARS. All we get are mumbo jumbo replies (if any at all) that barely half justify their total obstinance on this issue.
Over the years I have discovered several probably causes of the meta data clearing, caching. I recently discovered that if another IOS device on the network has infuse installed, an attempt initiated by it to update its meta data, even if no files or shares are connected generates an iCloud call, and the mismatch will crash the primary AppleTV Infuse meta data. EVEN if you have icloud caching disabled. !! Other reasons, if you have a hard drive on a share that’s gone offline, or drive letter lost, reassigned, it can cause meta data crashing and erasing. I have been begging for local storage both for convenience and privacy, but as of yet it’s been politely ignored. >fury< >stomp! >stomp!< >smash!< >smash!<
Other reasons, if you have a hard drive on a share that’s gone offline, or drive letter lost, reassigned, it can cause meta data crashing and erasing. I have been begging for local storage both for convenience and privacy, but as of yet it’s been politely ignored. >fury< >stomp! >stomp!< >smash!< >smash!<
Wow, I feel pretty ashamed I didn’t think of this. That’s the issue in my case!
I watch via SMB, my system runs 24/7. But of course I have to restart from time to time. Windows 10 re-assignes all hard disk ID’s (not the letter) randomly at restart, I have no idea why. That’s when the metadata cache cleares.
So yeah, local library / metadata on the server would be nice.
Very interesting. That’s the same case with me. I’m having a Mac and a NAS as sources. The nas is not always on.
@james Do you think that’s the reason?
I use an NAS (with a static IP assigned by my router based on its MAC address), and setup the share to connect to that IP via NFS. I haven’t experienced a database crash in ages (2 x 32 GB AppleTV 4Ks, each first-gen 4K).
Don’t see why a hard drive going off line should kill the database; Infuse is designed to work with sources that come and go.
I don’t download screensavers and got rid of the ones that were there by downloading some random large apps (to force them off) which I later deleted; and I’ve offloaded all but the few other apps I actually use (YouTube and YouTubeTV, plus no more than a couple of any other major streaming services’ apps at any given time). Those changes seemed to have been all that was necessary.
No, I don’t think so. The only time you will see this cache cleared message is when Infuse has been inactive and another app has requested more storage space, resulting in tvOS making room by clearing data from Infuse and other apps.
If you have a device which is offline, the metadata will remain intact.
However, if you have a device which is online but a particular drive is disconnected (or renamed) those items would be removed from the Library (the behavior is similar to if a file or folder is deleted). Once the drive is reconnected, then they would automatically reappear in Infuse. This however is a graceful action, and would not result in a metadata cache cleared message.
Yes I agree that is the way it should work, but I assure you I have seen the behaviors I have described. Also if too much memory is being used in the AppleTV by other apps creates metadata cache problems as well. You guys were totally denying the metadata call from my mobile Infuse caused the crash on my appletv even when icloud cache is disabled. But the immediate cause and effect was absolutely incontrovertible, It happened right in front of my eyes
FIrecore/Infuse developers should start taking our requests seriously for a pointer option in the menu we can enable to store metadata locally on premises. No way, no how can it be a difficult ask. The software is already written to look for an outside source. Just change the code to add a second location option. Have made the case pretty clearly early on regarding privacy concerns AND stability, The stubborn refusal to go forward with this is frankly bewildering and frustrating.
Getting really fed up with this.
That would be my suggestion, too. Always claiming, it’s the atvs fault, is pretty frustrating. Especially with so many different behaviours concerning the problem.
Why do similar devices and configurations get different results?
I most definitely do not write code for a living, but when envisioning all the different possible storage devices users might want to back up their data to, and all the different communication protocols, security schemes, and operating systems those devices might run, all of them historically released in a multitude of different versions, some regularly updated, some long neglected … and that it ought to be harder for software like Infuse to gain write permissions to my network hardware then it is for it to get read-only access … I think you might be underestimating exactly how simple (or not) a job that might be.
Certainly I’ve done that myself often enough.
Plus handling when multiple people are sharing a server.
Famous last words.
It would actually be a much more complicated task than it seems, with many edge cases issues that are certain to come up.
Instead, Infuse uses iCloud for backups and to keep things in sync. I understand there may be a few who prefer not to use iCloud for reasons, but for the other 99% it works well, allows devices to stay in sync, and dramatically shortens recovery times when space issues on the ATV do appear.
Still hoping for this someday
You guys are overthinking this. The Metadata Cache backup can can be incorporated as a folder/file/archive in an existing share/drive that Infuse is already looking at. There would be no need to search other parts of a network or change additional attributes or permissions. In fact it could just as easily be duplicated in case of disasters on other hard drives on the same share. I have several 8TB drives sitting in my SMB share. I too have done my share of coding, programming, setting up and administering networks since the 1980s Like I said, NOT A BIG ASK
You guys are overthinking this. The Metadata Cache backup can can be incorporated as a folder/file/archive in an existing share/drive that Infuse is already looking at. There would be no need to search other parts of a network or change additional attributes or permissions. In fact it could just as easily be duplicated in case of disasters on other hard drives on the same share. I have several 8TB drives sitting in my SMB share. I too have done my share of coding, programming, setting up and administering networks since the 1980s Like I said, NOT A BIG ASK
I have been having big problems with Infuse for the last several betas.
I am finding that the App seems to resets its activation status and claims that tvOS has cleared the Metadata cache because it is running out of space. This cannot be true, as I have an AppleTV 4K with 64GB of memory, and only 15 apps installed, and no games, just streaming services.
It happens nearly every day, and at the same time, Infuse needs to be activated again, as it asks me to upgrade to Pro. I’m used to this if I update the OS on my ATV, but this is far too often, and is highly annoying.
I’m using tvOS 16.5 beta 1. (This issue has been going on since tvOS16.3) Any ideas?
Sounds like maybe you should try completely resetting your Apple TV
Honestly said, I would not reset my atvs and set them up again, due to this permanent problem. This problem could be solved, the way it was discussed earlier.
That’s one of the reasons why I changed to a different system.
I have (3) Apple TVs:
- (1) ATV 4K 1st Gen (32GB)
- (1) ATV 4K 2nd Gen (64GB)
- (1) ATV 4K 3rd Gen (128GB)
Each added about a year apart from one another.
The Metadata cache clearing happens on all 3 but it happens most on the 32GB, and the least on the 128GB.
The other issue I experience, which to me is a much bigger issue, is the Apple Photos Screensaver (which I have set to Favorites album which has 12,000+ photos) stops working after presumably, the cached photos for the screensaver fill the entire disk space on the Apple TV. (More info on that: https://discussions.apple.com/thread/254254137 | https://www.reddit.com/r/appletv/comments/xlwhg9/screen_saver_only_playing_most_recent_photos/ ) - Eventually, instead of the Screensaver showing a random slideshow from my Favorite album, I just get a spinning wheel. Restarting Apple TV brings the screensaver back. When the Screensaver is working, it’s only limited to 100 photos or so at a time (as opposed to utilizing the entire 12,000+ photos). Like the metadata cache clearing, the spinning wheel problem with the Screensaver also happens most on the 32GB ATV and least on the 128GB ATV.
I bring up the Screensaver issue because I theorize it is related to the Infuse Metadata cache clearing issue and I believe the Metadata cache clearing happens most when I have the Photos Screensaver enabled.
While I don’t disagree with many of the users here that there may be some kind of solution Infuse could come up with, it does appear to me that Apple is not properly managing the Cache on Apple TV- even with their own built in Screensaver function.
Hoping there is a resolution at some point with both the Screensaver AND Infuse, both problems have persisted for a long time.
@stimpy88 said it was happening daily so my guess is something more significant is wrong. There are no permanent solutions, just workarounds. But you can almost guarantee issues with a 32 GB machine get 64 gig or more.