[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]
Re: parted offers existing encrypted PVs when making new VG
- From: Chuck Anderson <cra WPI EDU>
- To: Discussion of Development and Customization of the Red Hat Linux Installer <anaconda-devel-list redhat com>
- Subject: Re: parted offers existing encrypted PVs when making new VG
- Date: Mon, 17 Nov 2008 19:59:33 -0500
On Mon, Nov 17, 2008 at 03:47:16PM -0600, Dave Lehman wrote:
>
> On Mon, 2008-11-17 at 16:34 -0500, Chuck Anderson wrote:
> > anaconda-11.4.1.58
> >
> > I have sda5 encrypted PV with an existing VG on it. I want to create
> > a new VG to install to, so I select a free partition and click LVM.
> > In the list of Physical Volumes to Use, sda5 is checked off by
> > default, along with the new blank partition I want to use. This is
> > dangerous, because the user might unknowingly overwrite an existing
> > encrypted partition.
> >
> > This happens only when I enter the passphrase to unlock sda5. If I
> > don't enter the passphrase for sda5, I'm not offered that partition as
> > a choice for creating the new VG.
>
> Does the preexisting VG appear in the main partitioning UI? It should.
> If it doesn't, that means we didn't detect it, which in turn means the
> PV is unused and therefore available for use in a new VG.
Yes, the preexisting VG does appear if I give the passphrase to unlock
sda5. After some more investigation, I noticed that the encrypted PV
on sda5 actually has 1 free extent:
--- Physical volume ---
PV Name /dev/dm-0
VG Name l.sys
PV Size 27.68 GB / not usable 24.67 MB
Allocatable yes
PE Size (KByte) 32768
Total PE 885
Free PE 1
Allocated PE 884
PV UUID UwcvVJ-LeYA-xYg5-KFq9-W6F5-cEtQ-PPX5zi
So I guess technically this isn't a bug then, as long as it only uses
the free extents and doesn't overwrite the allocated extents. You
could create a new VG and use the 1 free extent on that PV along with
more extents on another PV.
I'm going allocate that last extent manually and try Anaconda again to
see what happens. If the behavior changes, then I think it is
behaving correctly. If it still offers to use sda5, then there is a
bug.
On the other hand, I believe Anaconda originally created this PV/VG
and the LVs on it--it was the one to leave that last extent free in
the first place. That's a different bug though.
[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]