From owner-freebsd-questions Tue Apr 11 7:45:11 2000 Delivered-To: freebsd-questions@freebsd.org Received: from smtp.visto.com (smtp.visto.com [206.79.140.96]) by hub.freebsd.org (Postfix) with ESMTP id 09EC837B785 for ; Tue, 11 Apr 2000 07:45:09 -0700 (PDT) (envelope-from jtelford@visto.com) Received: from mp7 (206.79.140.187) by smtp.visto.com (NPlex 4.0.058) id 38E3ECE0000BBD3F; Tue, 11 Apr 2000 07:43:54 -0700 Message-ID: <38E3ECE0000BBD3F@smtp.visto.com> (added by administrator@visto.com) Reply-To: jtelford@metroland.com From: "j telford" Subject: Re: Dump/Restore disk to disk trouble Date: Tue, 11 Apr 2000 07:44:52 -0700 X-Mailer: Visto To: cjclark@home.com Cc: questions@FreeBSD.ORG MIME-Version: 1.0 X-Mailer: Visto Server Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Mon, Apr 10, 2000 at 02:38:58PM -0700, j telford wrote: > First, I'm quite new at this. > 2 Seagate 8 GB SCSI Ultra Wide off an Adptec 2940 UW. > FeeBSD 3.4 >=20 > Complete new setup and no extra data on drives that didn't come with th= e=20 > install CD. >=20 > I realized I wanted to turn my /usr 16 GB slice into a vinum mirror to = the=20 > other drive so I thought I could use dump/restore commands first move /= usr to=20 a=20 > small 1 gb slice, create the vinum slice and copy it back. Also gives m= e good=20 > experience playing with dump/restore and vinum before turning users loo= se on=20 > the box. > Well and I'm done playing and now need some help please. > I searched back through some other posts and attempted the following: >=20 > Current /usr slice da0s1f > Create new slice on da1s1f > Boot to single user mode > fsck -y / > fsck -y /dev/da0s1f > fsck -y /dev/da1s1f > Clean on all >=20 > mount -u / > mount /dev/da1s1f /mnt >=20 > >From this point I tried 2 different dump/restore commands, both seem t= o have=20 > trouble with the restore portion. >=20 > Attempt 1 > dump -0f - /dev/da0s1f | (cd /mnt; restore -xf -) >=20 > We get past "Dump:dumping (Pass iv) "=20 > Then a multiple of ' cannot create file no such directory '= =20 errors. > Then a multiple of ' cannot create hard link no such direct= ory '=20 > errors. > Then : set owner . y/n: I answer y >=20 > /mnt now appears to contain some of the sub-directory tree but not all = of it. >=20 > Attempt 2 > dump -0f - /dev/da0s1f | restore -rf - > Gets past 'dump:dumping (Pass IV) again then after a few minutes then a= =20 > multitude of file names with the error 'unable to create device not=20 configured' > fopen: device not configured. > another pause with a message 'invalidating pack' > and finally 'cannot create savefile ./resoresymtable for symbol table' >=20 > I have also tried doing the dump and restore separately and can create = a dump=20 > file on disk but the restore from disk file runs for a bit then asks fo= r a=20 tape=20 > which I can't get around. Is /dev/da0s1f roughly the same size as /dev/da1s1f? What does, % df -i Produce when both are mounted? I'm just making a wild guess that those errors could be from running out of inodes on the target partition. --=20 Crist J. Clark cjclark@home.com Crist: /usr (/dev/da0s1f) was a 16 gb slice and the new slice was (/dev/da1s1f) = was=20 only 1 gb. So the inode difference was huge.=20 Today just for testing I made /dev/da1s1f 16 GB and had the same results = with=20 the above procedure. The dump/restore did not work. Any other thoughts ? harware timeout or something ? The errors fly by so = fast,=20 is there a log generated somewhere ? Thanks, John. ______________________________________________________________________ Get Visto.com! Private groups, event calendars, email, and much more. =20 Visto.com. Life on the Dot. Check it out @ http://www.visto.com/info To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message