Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 4 Dec 2004 02:03:42 +0000
From:      Adrian Wontroba <aw1@stade.co.uk>
To:        freebsd-stable@freebsd.org
Cc:        Ceri Davies <ceri@submonkey.net>
Subject:   Re: 4.x can't read 5.x dump?
Message-ID:  <20041204020341.B14096@titus.hanley.stade.co.uk>
In-Reply-To: <20041203143609.GL513@submonkey.net>; from ceri@submonkey.net on Fri, Dec 03, 2004 at 02:36:09PM %2B0000
References:  <200412020048.iB20mhiP009447@app.auscert.org.au> <20041202024052.GA11786@electra.cse.Buffalo.EDU> <20041203142415.GJ513@submonkey.net> <20041203142707.GD29117@electra.cse.Buffalo.EDU> <20041203143609.GL513@submonkey.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Dec 03, 2004 at 02:36:09PM +0000, Ceri Davies wrote:
> > > Should I expect a dump taken from 4.X to be restorable on 5.X though?
> > Yes.
> Phew.

I didn't even think about the possibility of dump not being forwards
compatible (8-(

In passing, you may find the buffer port useful.  I spent a while a
couple of years ago experimenting with settings, and find this sort of
thing sufficiently fast:

/sbin/dump -L -0 -u -C 32 -b 32 -f - /home |
	/usr/local/bin/buffer -s 32k -m 16m -t |
	gzip  | 
	/usr/local/bin/buffer -s 32k -m 16m -t > new/home.dump.gz

That is, big buffers and block sizes for dump, with a 16 MB buffer of
either side of gzip to absorb some of the delays in compression or
writing the dump file.  Sorry, no timings available at the moment.

Note:
* Remove the -L under Release 4
* When the above line is executed, the script has changed its CWD to the
  dump directory, usually on an NFS mount.
* When dumping to the same machine, I include -h 0 and make the dump
  directory (and a slew of other recoverable by other ways ones) nodump
  with chflags nodump.

-- 
Adrian Wontroba



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041204020341.B14096>