Mailinglist Archive: opensuse (2008 mails)

< Previous Next >
Re: [opensuse] Re: hard drive replacement question
  • From: peby@xxxxxxxxxxx
  • Date: Sat, 25 Apr 2009 11:51:56 -0400 (EDT)
  • Message-id: <41120.70.127.219.5.1240674716.squirrel@xxxxxxxxxxxxxxxx>

On Saturday, 2009-04-25 at 10:22 -0400, peby@xxxxxxxxxxx wrote:

David is totally correct about dd being the way to go - one slightly
better way to go though is ddrescue with does a much better job at
recovering the salvageable sections first, then retrying the bad sector
areas. You can also use a log file so the operation can be resumed if
interrupted, etc.

I have to disagree.

Not unless you have to do an fsck on the image, or if there are bad areas
which you try to recover. The old HD is still working, and there is a full
backup.

And even in that case, a file copy works fine, it will simply fail on
broken files, which can then be recovered from the backup instead.

Not to forget, there is clonezilla, specifically made for this situation:
replacing a HD, cloning a system.

- --
Cheers,
Carlos E. R.

Carlos,

Clonezilla is file system aware - and does not work on LVM, it simply
falls back to dd anyway in that case, and he might well be using LVM if he
used the default partitioning of a newer distro.

If there are bad sectors (which in this case it appears there are) using a
block copy method might be a better way to go and could potentially even
recover a read error on a sector where as rsync certainly will not. It is
also possibly rsync will simply fail depending on the number of bad
sectors, etc.

Either way will work, but why worry about partitioning, rsyncing, read
errors and recovery from backups when one command will likely accomplish
it all? ddrescue /dev/sda /dev/sdb. Done.

Pete



--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse+help@xxxxxxxxxxxx

< Previous Next >
References