Re: dealing with timestamp (in-)accuracy
[email protected] (Joerg Schilling) Wed, 12 Mar 2008 14:20:42 +0100
| Newsgroups | gmane.comp.archivers.star.user |
|---|---|
| Message-ID | <47d7d8aa.yDbdpR8HVjf5OwA8%[email protected]> |
Les Mikesell <[email protected]> wrote: > I thought the problem with direct dump-like access to a live filesystem = > had been known for ages: = > http://dump.sourceforge.net/isdumpdeprecated.html Correct, the problems are known at least since BSD in 1981 introduced the = dump code that was hacked for Linux. It is probably known even before when = there was a similar program for a UNIX-V7 filesystem that did restore to an = _unmounted_ target filesystem and retained inode numbers. > By the way - does anyone know how zfs and it's incremenatal send/receive = > compare? ZFS send/receive work at block level. This is very effective as the FS does= not = need to scan directory content but it does not allow to "restore" single fi= les. If you like to do backups that allow to restore single files, star is OK. L= ater = versions will add support for HFSv4 (NTFS) style ACLs. J=F6rg -- = EMail:[email protected] (home) J=F6rg Schilling D-13353 Be= rlin [email protected] (uni) = [email protected] (work) Blog: http://schily.blogspo= t.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/sch= ily