Ask HN: Your Amazon Glacier strategy with versioning and de-duplication on *nix?

3 points by balladeer ↗ HN
Command line, or GUI (preferred).

Something that just sits in the background, monitors change and keeps backing up the change (while taking care of the de-duplication).

What do you use? (I know Arq which I believe has been covered more than once here)

And how do you use it? What is your backup frequency? How do keep the cost minimal? Is it too frequent, or seldom - as in few snapshots a week?

Have you every had to restore? If yes, could you find a way to minimise the price of retrieval?

I've around 700GB of personal data that I just want to be sitting there in Glacier for the worst case scenario.

(PS. Yes, I've looked at Tarsnap and it doesn't work for me).

4 comments

[ 3.2 ms ] story [ 20.7 ms ] thread
Just out of curiosity, why does Tarsnap not work for you?
I am not a geek, nor some uber nerd. Tarsnap seems to be designed exactly for them. I once asked about a GUI of Tarsnap in one of the threads on HN few months ago and the very mention was heavily downvoted. So, I just think I don't fit in the fan-circle and honestly I am fine w/o it.

Another reason is price. I will put a good amount of data there lying unused for a long period so the extra dollars shall count a lot and since I'm searching for a client that would have de-duplication that one benefit would be settled too (as I've heard Tarsnap has de-dup).