|Subject:||Re: [rdiff-backup-users] Delete most recent increment?|
|Date:||Wed, 18 Dec 2013 16:56:26 +0000|
|User-agent:||Mozilla/5.0 (Windows NT 6.1; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0|
On 18/12/2013 15:43, Adrian Klaver wrote:
On 12/18/2013 03:29 AM, Ron Leach wrote:On 16/12/2013 15:28, Dominic Raferd wrote:Ron, see my utility here: http://www.timedicer.co.uk/programs/help/rdiff-backup-regress.sh.phpDominic, thank you for posting this. From its description, it would seem to do exactly what we need. I have not tried it. You'll think this odd, perhaps, but because this library is fairly important to us, I am reluctant to run anything that I could not first audit; for example, I would want to check for side effects, or 'assumptions' about the states of the source path or the backup path or the backups there or something, whether there was anything which might be problematic over NFS (file locking, for example) or whether the size of the increment might matter or, even, Debian-version sensitive, but these are just examples of concerns and is not meant to be an exhaustive list. Sadly, I'm not familiar with scripting language, so I was unable to check the script to see whether our data, or our backup regime would create any problems for, or incur any problems following, running the script. So, I'd prefer to do something 'manually', if it were possible, so that I could be aware of what changes were happening.Take a look at this; http://www.backupcentral.com/phpBB2/two-way-mirrors-of-external-mailing-lists-3/rdiff-backup-23/force-a-regression-of-rdiff-backup-archive-104028/
It was based on this insight of Janne's that I developed my bash script (my reply can be seen at http://lists.gnu.org/archive/html/rdiff-backup-users/2011-01/msg00030.html). The script has been slightly updated since that posting.
Robert's analysis is quite correct, the method is to trick rdiff-backup into thinking that the previous backup failed and thus make --check-destination-dir trigger a regression. He is also right to point out that regressions can take a long time.
I have used it with success, but I am not offering a warranty, so I quite understand Ron's caution.
Dominic -- *TimeDicer* <http://www.timedicer.co.uk>: Free File Recovery from Whenever
|[Prev in Thread]||Current Thread||[Next in Thread]|