That's metadata. The _data_ hasn't changed. I repeat it is wasteful
to back up *data* that has not changed. Some programs are smart
enough to do this.
I've also pointed out that we can have it both ways, if we want.
Apparently some people would rather back up the same identical data
over a
" <[EMAIL PROTECTED]>
> Subject: Re: [Bacula-users] checksums
> To: "'Michael Koppelman'" <[EMAIL PROTECTED]>,"'Dan Langille'"
> <[EMAIL PROTECTED]>
> Cc: bacula-users@lists.sourceforge.net
> Message-ID: <[EMAIL
o previous versions. We wouldn't care if they were
restored with the wrong mtime. In fact, I've never run into a
situation where I was concerned about the mtime of the restored files.
M.
On Nov 28, 2006, at 8:12 PM, Dan Langille wrote:
> On 28 Nov 2006 at 17:26, Michael Koppelm
Sorry if this is redundant but I just wanted to add my voice to the
mix that it is too bad that bacula backs up files that have not
changed just because their mtime changed. In the end, it is probably
less expensive to checksum than move and handle redundant data. It
would at least be nice
Thank you for your responses. I seem to have it working now. I set
the default pool to Full so that if any jobs get upgraded they will
go in the correct pool. The main confusion, which I already mentioned
but will reiterate is/was:
1. I would have thought or hoped that somehow the pool cou
OK, thanks anyway...
Michael
On Oct 3, 2005, at 8:49 PM, Michael Koppelman wrote:
I've been searching the archive and it looks like I'm not the only
one confused by pool behavior. I apologize if this is redundant but
I can't seem to find an authoritative answer.
My goal
I've been searching the archive and it looks like I'm not the only
one confused by pool behavior. I apologize if this is redundant but I
can't seem to find an authoritative answer.
My goal is simple: put all Full backups in the Full pool, put all
Differentials in the Differential pool and p