dump restore problem

[Matthew Dillon]

:On Sun, Apr 27, 2003 at 12:44:36AM +0300, Tomi Vainio - Sun Finland wrote:
:> During my ufs2 migration I've used piped dump restore procedure.  Is
:> it normal that block sizes like 512 or 1000 don't work?  Block size
:> 512 gives an error and 1000 hangs forever while 126 is working fine.
:> 
:> dump 0buf 512 - / | restore xbf 512 -
:
:I think you want to leave out the b option and use -a to tell dump not
:to try to determine the size of the "tape".  I would also use -rf on
:restore, not -xf, and add -C 32 to dump args.  At least that's what
:I gleaned from rtfm dump & restore.
:
:-- 
:Barney Wolff         http://www.databus.com/bwresume.pdf
:I'm available by contract or FT, in the NYC metro area or via the 'Net.

    The block size is too large, use 127 (I use 64 myself).  Remember,
    the block size is in 1K chunks so 127 = 127KB, which is plenty big
    enough.  And definitely use 'a'uto sizing in the dump command.

    Generally speaking this is what I do:

	dump 0abf 64 - / | restore rbf 64 -

    Don't use 'u' in the dump command unless you are making a permanent
    backup that you want the dump system to record in /etc/dumpdates.
    If you are just using dump/restore to copy a filesystem, do not use
    'u'.

    Use restore's 'r' option instead of 'x' if you want an exact restore.
    'x' is used when restoring part of a backup, 'r' is used when restoring
    the entire backup.  For example, if you just wanted to restore the 'etc
    directory you would use:

	dump 0abf 64 - / | restore xbf 64 - etc

    But if you want to restore the whole thing you would use:

	dump 0abf 64 - / | restore rbf 64 -

					-Matt
					Matthew Dillon 
					<dillon at backplane.com>