From: Jeff Macdonald <macfisherman@gmail.com>
To: "M. Matt Colgin" <mcolgin@gmail.com>
Cc: LVM general discussion and development <linux-lvm@redhat.com>
Subject: Re: [linux-lvm] Adding a disk to a VG gone bad - need help
Date: Mon, 28 Feb 2005 21:21:58 -0500 [thread overview]
Message-ID: <45ae90370502281821514a77c2@mail.gmail.com> (raw)
In-Reply-To: <a58063aa0502281019185c9845@mail.gmail.com>
On Mon, 28 Feb 2005 10:19:46 -0800, M. Matt Colgin <mcolgin@gmail.com> wrote:
> Weird, I'm not totally following how/what happened, but it sounds like
> you VG went from 1 PV to 3 (but maybe you started off with two).
Yes, I started with 2.
> Hopefully you don't have some super important data on it. If you do,
> there are some people with more knowledge than me, repost your
> question with some more/updated information.
>
Just all my digial media! :(
> Each LVM command will write a "changelog" to the /etc/lvm/archive/*
> directory with the state of the VG before/after the command was
> issued. The best bet to undo the extend would be to issue a
> vgcfgrestore with the file, before your vgextend.
Well, with Fedora Core 3, this isn't possible, because that data was
stored on a LV. They should make it a symlink to /boot like grub is.
<snip stuff about cables>
This was a disk that was being used as a normal non LVM disk. So it
was properly working while copying data from it to a LV. After all the
data was transferred, I wanted to add it to the existing VG. See
procedures I followed from the previous mail below.
As I outlined before, it seems that doing:
vgreduce --removemissing VolGroup00
may do the trick, but in test mode I see this troubling line:
Removing LV Home from VG
Which is definately not what I want. However, perhaps rerunning
pvcreate -u with the UUID it claims it doesn't have a device for is a
viable alternative?
> On Fri, 25 Feb 2005 23:22:59 -0500, Jeff Macdonald
> <macfisherman@gmail.com> wrote:
> > Hi,
> > I have one volume group and that is used as such:
> > [root@jeff ~]# /usr/sbin/lvscan
> > ACTIVE '/dev/VolGroup00/Root' [9.78 GB] inherit
> > ACTIVE '/dev/VolGroup00/Home' [231.31 GB] inherit
> > ACTIVE '/dev/VolGroup00/Swap' [320.00 MB] inherit
> >
> > I added a whole disk to VolGroup00 like so:
> > [root@jeff ~]# dd if=/dev/zero of=/dev/hde bs=512 count=1
> > 1+0 records in
> > 1+0 records out
> >
> > created the physical volume:
> > [root@jeff ~]# /usr/sbin/pvcreate /dev/hde
> >
> > added it to the volume group:
> > [root@jeff ~]# /usr/sbin/vgextend VolGroup00 /dev/hde
> >
> > added 100G from the 150G drive to Home:
> > [root@jeff ~]# /usr/sbin/lvextend -L+100G /dev/VolGroup00/Home
> > Extending logical volume Home to 331.31 GB
> > Logical volume Home successfully resized
> >
> > At this point the docs suggest umounting the logical volume in order
> > to resize the file system. So I decided I was going to reboot the
> > computer and start it into single user mode. When doing that I'm
> > greated with:
> >
> > lvm exited abnormally!
> > Couldn't find device with uuid <uuid>
> > Couldn't find all physical volumes for volume group VolGroup00
> >
> > So it seems that it can't find the new physical volume that I just
> > added. LVM is totally new to me, so any pointers would be great.
> > Booting from CD into rescue mode I wasn't unable to get very far. I
> > was able to run lvm and pvscan shows this interesting line:
> >
> > PV /dev/hdf1 VG VolGroup00 lvm2 [ X GB / X GB free]
> > PV /dev/hdg2 VG VolGroup00 lvm2 [ Y GB / Y GB free]
> > PV unknown device VG VolGroup00 lvm2 [ Z GB / Z GB free]
> > Total: 3 [ X GB ] / in use 3
> >
> > It seems to be that vgreduce is what I want in order to start over,
> > but running this in test mode with -t and --removemissing I see this
> > worrying line:
> > Removing LV Home from VG
> >
> > and the line I want
> > Removing PV with UUID {correct UUID}
> >
> > Am I on the right track?
> >
> > TIA
--
Jeff Macdonald
Ayer, MA
prev parent reply other threads:[~2005-03-01 2:22 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-02-26 4:22 [linux-lvm] Adding a disk to a VG gone bad - need help Jeff Macdonald
2005-02-28 18:19 ` M. Matt Colgin
2005-03-01 2:21 ` Jeff Macdonald [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=45ae90370502281821514a77c2@mail.gmail.com \
--to=macfisherman@gmail.com \
--cc=linux-lvm@redhat.com \
--cc=mcolgin@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox