From: Peter W. <pw...@it...> - 2009-10-16 16:14:51
|
Robert Kosinski wrote: > If the data is non-overlapping, restore the "dead data" to the live > laptop, allow a backup to happen, then delete the obsolete sets. If > you can't figure out which data is unique then the backups probably > aren't of much use anyway. By the time you need them, you really won't > know what's going on. > > On Fri, Oct 16, 2009 at 11:49 AM, Omid <bac...@om...> wrote: > >> i think that on one hand, this may not matter, because pooling would make >> sure that files are only stored once. >> but it does make a bit of a mess in so far as having one machine refer to >> one thing goes. >> i'd like to do something similar (the machines on one network are very >> poorly named, and i didn't want to change all the names around, but didn't >> think of using clientalias's, which i should have. so now we have machines >> named corp-1, corp-2, corp-2, corp-xxyz, and it's a PAIN to know what >> belongs to who!!!) >> thanks. >> >> On Fri, Oct 16, 2009 at 8:24 AM, Peter Walter <pw...@it...> wrote: >> >>> I have a situation where a single laptop was replaced twice, but with >>> all the data ghosted from one laptop to another. The computer name was >>> then changed, and a new BackupPC entry created. I therefore have three >>> sets of non-overlapping backups in BackupPC, but they really refer to >>> the same set of data. How can I merge the three sets of backup data into >>> one set going forward? >>> >>> Peter. >>> The backups are what are non-overlapping, not the data, meaning, each backup set contains full and incremental backups where the dates of the backups do not overlap with the dates of the full and incremental backups in the other two sets. |