Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 24 Apr 2004 12:06:43 -0400
From:      Richard Coleman <richardcoleman@mindspring.com>
To:        Don Lewis <truckman@FreeBSD.org>
Cc:        julian@elischer.org
Subject:   Re: Testing Tar (was Re: bad news for bsdtar..)
Message-ID:  <408A9093.2050409@mindspring.com>
In-Reply-To: <200404240540.i3O5eA7E053079@gw.catspoiler.org>
References:  <200404240540.i3O5eA7E053079@gw.catspoiler.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Don Lewis wrote:
>>>At least the -current version of tar skips reading the
>>>data when it is writing to /dev/null.
>>
>>A-ha!  That explains a few of the odd timings I've seen.
>>I wonder why it does that?  (Other than to look good on
>>benchmarks, of course. ;-)
> 
> 
> This speeds up Amanda quite a bit.  Amanda will run tar with the
> --totals option as well as other options to specify either full or
> incremental backups multiple times for each file system that it backs
> up.  It does this to plan the best mixture of full and incremental
> backups.  If tar actually read the data from disk each time, the
> planning phase would take a *lot* longer, and would thrash the disk a
> lot more.

Until libarchive gets support for sparse files, it's probably better to 
stick with gtar or rdump with Amanda.

But the concept of a version of Amanda that natively uses libarchive is 
very cool.  It seems like a natural target.

Richard Coleman
richardcoleman@mindspring.com



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?408A9093.2050409>