*BSD News Article 49424


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!fu-berlin.de!zib-berlin.de!irz401!uriah.heep!bonnie.heep!not-for-mail
From: j@bonnie.heep.sax.de (J Wunsch)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: Backing up the whole 9 yards...???
Date: 21 Aug 1995 12:10:55 +0200
Organization: Private U**x site, Dresden.
Lines: 26
Message-ID: <419m3f$q6d@bonnie.tcd-dresden.de>
References: <DDHIzE.25r@agora.rdrop.com> <41198g$mb@gate.sinica.edu.tw> <415rlh$na@palmer.demon.co.uk>
Reply-To: joerg_wunsch@uriah.heep.sax.de
NNTP-Posting-Host: 192.109.108.139
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit

Gary Palmer <gary@palmer.demon.co.uk> wrote:

>Anyone know how well FreeBSD's dump command handles active (or
>hyperactive :-) ) filesystems?

I think it does.  It performs a `tag' run at the beginning, and tries
to find out which files to dump.  The dumpdate as recorded in
/etc/dumpdates (for the next incrmental dump) is that of the beginning
of this tagging operation, so files that have been changed between the
start of the dump and the time where they have been tagged might be
included in two dumps.

Of course, if a file disappears between its tagging and the actual
time when it should be dumped, dump will be at a loss. :)

If a file is modified after tagging, it will not be included in this
dump (but in the next increment).

I'm regularly using dump on active file systems, despite of the
potential race conditions.  I've been impressed by the various
possibilities `restore' is providing...
-- 
cheers, J"org                      private:   joerg_wunsch@uriah.heep.sax.de
                                   http://www.sax.de/~joerg/

Never trust an operating system you don't have sources for. ;-)