Re: [Duplicity-talk] are periodic full backups necessary?

[Olivier Croquette]

Peter Schuller wrote, On 20/01/08 7:57:
> Perhaps a compromise is possible whereby each individual file being backed up 
> is separately backed up and encrypted, such that the rsync algorithm can be 
> applied even with an untrusted remote system on the encrypted data. However 
> this also has security implications since you can make more determinations on 
> the number of files, their sizes, the distribution of changes over time and 
> so on than you can do with the volume uploads.

I thought about this exact approach with the same conclusions.
You have to make a compromise between confidentiality and reuse 
potential of the backed up data.

> Perhaps a single large "virtual" volume could be generated for a complete 
> backup, would would then be used to apply the rsync algorithm to the previous 
> volume. This assumes the rsync algorithm only requires one pass (does it?) 
> and that it will all work well even in the face of large displacement of data 
> in this huge file (probably not).

OK, I must admit I never thought of that one :)
I don't think it will play well with the encryption.

> Anyone have better ideas?

Another idea might be to use an encrypted virtual volume and to use hard 
links in it, like some tools do.
With hard links, it's pretty easy to implement backward diffs.
And confidentiality is ensured, since it's only a huge file.

But I am not sure how 1. trustworthy and 2. portable a virtual encrypted 
system can be.
Especially 1. may be problem if the link is not reliable.

Other issues to solve would be the size of the image, which can become 
pretty big, and dynamic growing thereof.