This is one way:
/bin/csh
setenv TAPE /dev/rmt/0hn (of whatever)
/usr/sbin/ufsdump 0ufbsd $TAPE 64 100000 100000 <file system>
** OR **
You can make a script for this, using a list for filesys
so that you can run multiple file systems.
ex:
#!/usr/bin/csh
setenv TAPE /dev/rmt/0hn
echo "Checking tape status ..."
echo ""
mt status
if ( $status != 0 ) then
exit 1
endif
set count=0
date >> tape_info
echo "using $TAPE" >> tape_infodate
# SET TO BEGINING OF TAPE
/usr/bin/mt rew
foreach filesys ( / /usr )
echo ""
echo dumping $filesys to $TAPE#
/usr/sbin/ufsdump 0ufbsd $TAPE 64 100000 100000 $filesys >>
tape_info
echo "" >> tape_info
echo ""
end
echo "" >> tape_info
/usr/bin/mt rew
/usr/bin/mt offline
echo dump finished date
echo ""
Scott Sanders wrote:
>
> So assuming I know what tape has a level 0 of my root file system, what might
> the commands look like to retstore that file system. I'm running Solaris and
> as I said use tar for the backups. I don't need specifics as I know a lot of
> people on the list are running Linux but the commands in general would be
> great!
>
> Thanks
>
> Joshua Baker-LePain wrote:
>
> > On Tue, 23 Jul 2002 at 9:13am, Scott Sanders wrote
> >
> > > I've plowed through the install and setup of amanda, thanks in no small
> > > part to the folks on this list, and have successfully restored various
> > > file systems with no problem. But the one task that is always the most
> > > daughnting (sp?) is restoring root from a total crash. For a client
> > > machine I'm assuming I'll need to reload the client OS and amanda but
> > > I'm not sure where to go from there.
> >
> > Client OS -- yes. amanda -- not necessarily. You can get the data off
> > amanda tapes using mt, dd, and tar, whch is rather handy.
> >
> > > The next big question is how do you recover from the amanda server
> > > crash?
> >
> > Each night as part of my backup script I tar up all the amanda config and
> > history info and put it in /home/amanda (backed up) as well as copy it to
> > a NFS mounted RAID (also backed up). That way I have all the history info
> > available in multiple places (and in multiple tape sets) should the worst
> > happen.
> >
> > Should the server die, it's simply a matter of reinstalling the OS,
> > restoring that tarball, and I'm back up and running.
> >
> > --
> > Joshua Baker-LePain
> > Department of Biomedical Engineering
> > Duke University
>
> --
> Scott Sanders
>
> Systems Administrator
> Concepts Direct, Inc.
> 2950 Colorful Ave.
> Longmont, CO 80504
