Ask Ubuntu is a question and answer site for Ubuntu users and developers. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I’ve started using rsnapshot as my backup system for my home PC. I really like the idea of hard links and how they are handled. But I can’t find the best workflow. Currently I keep my snapshots on the same partition and will copy the newest snapshot to a pen-drive at the end of the week.

Cloud storage is what I’m looking for. Dropbox doesn’t fit my needs, because there is no way to make Dropbox respect hard links — all snapshots are treated as full snapshots.

Renting a server is pretty expensive, so my question is, are there better alternatives for backup in the cloud? I would like to benefit from hard links and send only incremental backups, just like I do with my local host.

share|improve this question

You run rsnapshot on the remote server, it rotates the snapshots using hard links and rsyncs your files from your local machine over ssh.

rsync only sends over the changes so your bandwidth use is minimised, rsnapshot on the server manages the hardlinks to minimise disk space.

What the best cloud solution is depends upon how much disk space you need and other things

share|improve this answer

Symlinking inside cloud storage is just going to create a copy, not a link, so you're only going to benefit from rsnapshot if you tar the target directory before uploading it.

I admit, I haven't tried this before. You'd have to tar the file in a way that tar's the hardlink as a hardlink and not the file itself (or you'll just end up with another full backup, only compressed).

I'd take a look at man tar and see if it can be done. For example, it appears that the --hard-dereference (follow hard links; archive and dump the files they refer to), is off by default, so as long as you don't specify it, then your tar dump of the latest rsnapshot should be incremental.

share|improve this answer
I read manual again and it says that snapshot_root must be a local file system. So I suppose that it is not possible to make a backup from localhost and send it over ssh to a remote and make rsnapshot to deal with hard links. Hm, it is rather a remote that should pull my data(?). I think it makes sense. What do you think? – user10397 Feb 8 '11 at 17:22
That's kind of what I was getting at by suggesting tar'ing the resultant snapshot directory, then sending that to the remote. I'm not sure how you'd set up rsnapshot to grab directory from a remote site - I think you'll be beyond the remit of the tool here. Rsnapshot is brilliant, but quite a basic tool. – Scaine Feb 8 '11 at 18:58

I would consider using duplicity. It does basically this, by making TAR backup files and uploading to WebDAV or CloudFiles (SWIFT-style API) systems.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.