Size does not matter. The cost of running a snapshot is only a few seconds of CPU-time after the initial snapshot. How many seconds does it take to snapshot all of it if I may ask?
Do the contents of these huge folders change a lot in between your desired schedules? If so you might want to list these paths as separate snapshots with custom policies (e.g. per user@host:/path. In this example snapshots are taken every 4 hours. Once a month would ammount to 2600000 seconds (30x24x60x60 and rounded to the tune of Captain Crunch ;).
Sorry, perhaps I’m not understanding how this program works.
I would like to back up to B2 backblaze.
I might have faster internet than I used to, but if I have content which is important to me but not critically amazing important it’s right up to date, I would like to dictate the frequency it uploads tens of gigabytes of data on a metrered connection.
Further more, that reminds me, where is the option to throttle upload speed during waking hours?
First of all, and to you primary question, finer grained control of scheduling snapshots is actively being worked on and - given the pace of development - should be in in one of the next releases (if not the next release). This is a feature in a lot of demand.
Second of all, kopia might indeed be smarter than you are currently aware of. The rolling hash feature prevents backing up your 64GB file fully every time a snapshot is made, only the diff is snapshotted
From the features page " Kopia also uses splitting based on rolling hash, which allows efficient handling of changes to very large files. For example a virtual disk image which gets modified can also be efficiently snapshotted by only uploading the changed parts and not the entire file."
And now for your two cent. Uploading to backblaze is free of charge (class A-transaction) and storage is calculated on a per month basis. So if you’d snapshot your large file(s), change them a bit and snapshot again immediately you would see if all works to your liking in a day or two, and for your two cents.
Thank you, I appreciate it very much. The program seems particularly good compared to many other offerings.
So I’ve realised, I’ve misled you in my situation and I just want to clarify it, how it’s impacting me and may impact others. (Hence the desire for scheduling)
So now that I finally have time to reply and think about it, the issue is, in my case, a file is being written every 3 days, which is encrypted and 30 to 60GB in size, right?
Now the problem is, the file name, literally changes, per backup. It’s a completely different filename. The old file is bzzt gone.
So again, even though the file on the servers, may be several weeks out of date, as long as I know “at least one, full disk backup, less than 1 month old” is on the system - I’m happy.
So for at least one directory (or, sadly, a new custom backup job?) I would like to specify "hey, like don’t even snapshot, don’t bother, with engaging with the content in this particular folder, except once per month)
Thanks and sorry for not being clear. (FWIW the tech you’re describing sounds great, I recall diff patches for Doom, many, many years ago)
So, I’ve come back and re-read this post and I think this is actually what I’m after all along? ! (apologies! thank you)
So I guess the only improvement would be for simpler users, perhaps having buttons or sliders for days / weeks / hours and what have you. Though I think I can operate a calculator and work out what I’m trying to do.
Now my other post which was related to this, bandwidth control / throttling and possible scheduling of when backups upload (Regardless of snapshot time) might be quite useful!
However, I think with a bit more careful reading I’ve realised this should cover the job, thank you.