When you sync certain parts to a cloud services. It is not truly a backup. It adds the value of being offsite, which is very important for data, and accounts for most hardware failures. But if you’re house burnt down, or one of the many other things that could destroy everything at one location. (It’s fine to do this yourself. But it’s a pain to keep multiple off-site places to store your data, and sync them.)
When you use the sync feature of many of these services. It’s not a backup because if you accidentally change or delete anything, by accident, error, malware, etc. Those changes get pushed out, and you have no good copy.
Me, I just leave folders like ~/Dropbox, and whatever else, in my backups, and just use both without doing any extra work. (Local storage is cheap compared to cloud storage, and it’s only keeping the changes between snapshots.)
I am mainly using pCloud.
If I accidentally delete a file it will be in the trash bin of the cloud where I can restore. I believe some services have an option to roll back the same file to an earlier version, I know this feature for google sheets and google docs.
If it is modified somehow by malware or so then I don’t know . There is always risk.
But I am on Linux, what malware or viruses you are talking about!
Edit: @Kobold I know what you are talking about. I believe pCloud will fit your bill for privacy. AFAIK they store your data encrypted, so even pCloud themselves can’t read your data.
i remember they have an option to encrypt your files localy on your computer before even being transmitted.
You can review and tell me what you think.
Give it a try and tell me what you think.
I am using pCloud and for me it is the best cloud storage out there, I see it better than Google Drive and Dropbox. Google drive is hard to integrate with KDE, Dolphin… etc. and Drop Box limits me to like only 3 machines. If I need more I have to pay!
Honestly, if you sign using this link upon sign up you will get 1 extra GB of storage for free, and I will get the same. So it is better than just going to the website on your own and signup, you won’t get this extra 1 GB free. You can get free cloud storage up to 10 GB by referring your friends or inviting them to signup.
I would be happy to help create a wiki page that describes and explains the manual btrfs rollback.
I have already done this myself, but I am hesitant about the wiki article because my own systems work with a special btrfs layout. That’s why I could end up with commands that are tailored exactly to my btrfs layout and don’t work for others.
In addition, for security reasons, it is always necessary that at least 2 other people test it on their own.
If you would like to tackle this with me, I would suggest the following approach:
Create a new thread and move the relevant posts there.
Agree/discuss the procedure in this thread
Create a wiki page called “manual rollback with btrfs” or something similar (I have wiki write permissions)
Improve the wiki page until all steps always work regardless of the btrfs layout
Consider whether implementing it in a script is feasible and advantageous (risk assessment)
Possibly create a script for the rollback and incorporate it into the wiki
That would be great! For non advanced newbies. For me I am done already.
Though I wish a few bash script to run them in order because there a re a few reboots during the process, and a few reboots to test it is working!
I won’t say it was difficult, but way too many for an impatient old man like me!
I mentioned I followed steps I found on EndeavourOS forums.
Maybe you can check the forum.
The steps were simple, but too many for me. But they are like in 2 or 3 threads mainly.
I left my main workstation (and host) on Timeshift, but I do most tinkering in guests. Where I can purposely screw things up. Or even have both Snapper and Timeshift running, which is fine. (So I can have any layout.)
Can do. But is this the only way to start it. Public forums?
Which one, or both? (I actually just tested a wiki edit.)
I envision a method that would work with restoring any snapshot. Whether it’s from Snapper, Timeshift, or just running btrfs sub create yourself.
This I would really like to avoid. There are just too many scenarios to account for, and these steps are simple, but need to be explained in detail. Being software agnostic is a big plus too. The script would need to become pretty advanced, to the point that you may as well make your own TimeSnapper in a real language. Oh, wait you did that.
And you already have one person ready to run this script, without even knowing it’s going to do!
Two years ago, I got hung up on things like, “Where’s btrfs sub rename!?” Why does btrfs prop set <subvol> not work!?
It would of been nice to see all these methods of doing each step in one place. I had quite a few, “Is it really that simple?” moments, then there are not apparent things. And I had to sift through many sources to find them.
. There is always risk.
