Hi, Martin Leben wrote on 2008-07-29 16:00:04 +0200 [Re: [BackupPC-users] Explanation of archiving process]: > Joanne Cook wrote: > > The documentation says that the Archive function uses TarCreate, which > > merges > > incremental backups automatically. Does that mean that if you create an > > archive of a given incremental backup you are actually creating an archive > > of > > that increment plus the previous full backup? > > Yes. Backuppc will make sure that the created archive contains (at least) all > the files that were present on the client when the incremental was made. > > If the backups were made using transfer method rsync or rsyncd the archive > would > not even contain files that were deleted between the last full and the > incremental. (Someone with more experience, please correct me if I am wrong.)
that is correct, but sounds complicated (so does my version, sorry :). The archives will approximate the state of the data set you are backing up at the time the backup was taken. Incremental backups are an optimization, a "faster way to take a snapshot", but it's intended to be the same snapshot as for a full backup. A tradeoff of tar and smb incrementals is that they cannot detect file deletions, renames, or creation of files with timestamps in the past (before the reference backup), so, in these respects, the approximation will be inexact. This is true for both archives and restore operations. Aside from that, the implementation details of full and incremental backups should be invisible to the user. > > And to restore from a given > > incremental all I would need is that archive and not the previous full > > backup? > > Yes. > > > Presumably also if I create nightly archives of the previous incremental > > backup they are going to get increasingly large? > > Syntax error in your sentence...! ;-) Missing comma? But since we're splitting hairs ... > Did you meant to ask "Will the nightly > archives I create from incrementals get increasingly larger?". Yes they > will, > unless you are using transfer method rsync or rsyncd. (See note above.) With the exceptions noted above for tar/smb incrementals, the archive size should reflect the size of your data set. If your data set grows, so will the archives. If your data set shrinks, so will the archives (except that deleted files may still be present, adding to the size). But shrinking *files* may well lead to shrinking archives even for tar/smb incrementals (suppose you truncated large log files to 0 bytes instead of deleting them). Normally, your archives will probably grow, but it's *not* as though you needed <space for full backup> + <space for delta day 1> + ... + <space for delta day N> for the archive of an incremental backup N days after the full backup. It's more like <space for full backup>, regardless of whether a full or an incremental was actually taken on that day. Regards, Holger ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ BackupPC-users mailing list BackupPC-users@lists.sourceforge.net List: https://lists.sourceforge.net/lists/listinfo/backuppc-users Wiki: http://backuppc.wiki.sourceforge.net Project: http://backuppc.sourceforge.net/
