That would not be true resumability, but checkpointing (I know, a nit).
You are correct. That could be done every hour or so and would make
restart easier.
...Thanks,
...Ken
address@hidden wrote:
In case resumability gets implemented someday:
Given the fact that a resumed backup can be seen as a string of
incrementals. The first full request was interrupted now only missing
(or inbetween changed) parts will be transferred.
Viewing at the problem from this angel there seems to be a kind of easy
route in implementation.
At specific control points a temporary manifest (it holds the file meta
data list right?) has to be uploaded. It indicates that a backup was
underway and which data is already uploaded. The next backup routine
finds it and can resume from the last manifest up to a final manifest.
Hope this makes sense... just wanted to write it down before I forget it
.. regards ede
--
Hello
Kenneth Loafman wrote:
Duplicity does not have a resume capability. The suggested route is to
subdivide the backup into manageable chunks, said chunks being whatever
size you are willing to redo. My general approach is to make each
backup less than about 4 hours.
FWIW, my monthly full backup takes about 10 hours (for 185GB of data)
and I did not have any major problem so far (*knocking wood*), only
difference is that I use FTP and not SCP...
a+
Nicolas
_______________________________________________
Duplicity-talk mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/duplicity-talk
------------------------------------------------------------------------
_______________________________________________
Duplicity-talk mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/duplicity-talk