this post was submitted on 13 Jan 2026
12 points (100.0% liked)

Linux Questions

3317 readers
19 users here now

Linux questions Rules (in addition of the Lemmy.zip rules)

Tips for giving and receiving help

Any rule violations will result in disciplinary actions

founded 2 years ago
MODERATORS
 

Last year I bought an M-Disc drive and a bunch of M-Discs and burnt all my files in Nextcloud to them as a backup.

It's a new year, I want to take any files added or modified in 2025 to burn a new disc.

I can work out how to get all files with a modified date in 2025, the problem is that if a file was moved into Nextcloud but wasn't changed, the modified date doesn't change. So I will miss files that someone has moved into Nextcloud without changing if the modified date is before 2025.

I can't use created date as literally all the files have a created date 1 Jan 2025 or later as the created date is when they synced from the server.

Normally I'd rsync to find changes but the current copy is spread across like 10 M-Discs, and reading each of those 100GB discs at CD reading speeds is going to be painful.

Does anyone have a better idea?

Edit: In case anyone is finding this later, I didn't get a better plan other than planning ahead. I copied each disc onto my hard drive (into a folder "Old replica"), copied the current state into a different directory ("New replica"), and ran Czkawka to remove files from "New replica" that are duplicates of files in "Old replica". And also used the Czkawka setting to delete empty directories once I deleted all those files.

What was left in "New replica" I burnt to disc, then Rsynced these back into the "Old replica" which I have for now left on my hard drive. Maybe if I run out of space I'll consider saving hashes or something for comparing, but for now this is just an extra copy I am storing because I didn't find a better way 🤷

you are viewing a single comment's thread
view the rest of the comments
[–] Dave@lemmy.nz 1 points 2 weeks ago

I've had a play in the database. I can find the files and folder structure, then there are version timestamps which are probably what I'm looking for. But marrying these up to actual files for copying is going to be messy. At this point I think I'm just going to spent the days copying data off disks into a record of the last archive, then keep this on disk for next year. Or once I have a reconstructed archive, I might be able to build an index from it.

I didn't really think about this when burning the original discs (other than "that's a problem for future Dave"), so I have folders plopped on discs that are not in the same structure as in Nextcloud, as I spread things around to get them to fit. Once I reconstruct a proper version and get up to date, I can probably build some sort of index for next time where if a file has changed or it's not in my index, then include it.