Re: [Duplicity-talk] Incremental Backups always against the last full snapshot

[edgar . soldin]

On 17.12.2011 13:06, Daniel Weigl wrote:
> Hello List,
> 
> As far as I understand, incremental Backups save the differences against the 
> last _incremental_ snapshot (or full, if no incremental is available). At 
> least as I have just tested, a restore downloaded all incrementals until the 
> last full one.
> 

right

> Wouldn't it be good, to have a option to set duplicity to calculate all 
> incrementals against the last full snapshot. I know that these means the 
> incrementals will grow over the time, but on the other hand restores would be 
> much faster.
> 
> So depending on the nature of the backup'd data and restore-requriements (who 
> wants to wait long while the server is restoring?), i think this might a good 
> option.
> (off course only optional, and default is as it is now)

nifty idea, this could also serve to shorten a chain by making incrementals up 
to the new full incremental obsolete and deletable. the m.o. you suggest is 
surely an alternative for people with few changes in their backup sets.
 
> Is there a technical reason which might block this idea?

duplicity currently assumes that incrementals are based on the backup set 
before the incremental, hence these "full-incrementals" would have to be tagged 
for duplicity to detect that they are based on the full set. the backend 
volumes are named like 

duplicity-(full|inc).DATETIME.(manifest|volX.difftar).gpg

you'd probably want to exchange full/inc in the name for these. of course the 
purge routines and others will have to be modified accordingly too then.

but generally speaking, no show stopper here.

ede/duply.net

PS: what is this in your signature below?

> ++++++++[-<++++>>++++>+<<]>>++>++<<<<[- [->+<]>[->.<<+>]>>>[-[->+<]+>[<+
> +++++++[->++++++<] Daniel       >-.-[-<++>]<.[-]]++<[->-<]++>[<->+++++++
[->++++<]>..[-]<]>>]+<<<[-[->+<]+>[-<+>>>-[->+<]++>[-<->]<<<]<<<<]>>.<<<]