Bug 614812

+++ This bug was initially created as a clone of Bug #593642 +++

When an inconsistent lvm vg is found the user can choose to ignore the involved pvs (often partitions), these then get added to anaconda's immutableDevices list, yet autopartitioning with a clearpart type of linux or all will still remove the
involved partitions as shouldClear() does not honour the immutableDevices list.

--- Additional comment from pm-rhel on 2010-05-19 14:55:06 EEST ---

This request was evaluated by Red Hat Product Management for inclusion in a Red
Hat Enterprise Linux major release.  Product Management has requested further
review of this request by Red Hat Engineering, for potential inclusion in a Red
Hat Enterprise Linux Major release.  This request is not yet committed for
inclusion.

--- Additional comment from hdegoede on 2010-05-19 18:09:46 EEST ---

Note to QA. To reproduce this, do a default autopart install on 2 disks, then remove one disk (easiest with a vm), and start a new interactive install when prompted about the imcomplete volgroup, choose ignore, then choose clear linux partitions, check review partition layout and continue. You should get a complaint that there is not enough free space, instead the pv of the broken vg will get re-used, which should not happen as it should be ignored.

Can we get a qa-ack for this please ?

--- Additional comment from hdegoede on 2010-05-20 12:53:45 EEST ---

Fixed in anaconda-13.21.44-1, moving to modified.

--- Additional comment from releng-rhel on 2010-05-21 15:19:19 EEST ---

anaconda-13.21.44-1.el6 included in compose 'RHEL6.0-20100520.0' and is marked as having fixed this bug.
Moving to ON_QA.

--- Additional comment from atodorov on 2010-07-14 16:14:09 EEST ---

(In reply to comment #2)
> Note to QA. To reproduce this, do a default autopart install on 2 disks, then
> remove one disk (easiest with a vm), and start a new interactive install when
> prompted about the imcomplete volgroup, choose ignore, then choose clear linux
> partitions, check review partition layout and continue. You should get a
> complaint that there is not enough free space, instead the pv of the broken vg
> will get re-used, which should not happen as it should be ignored.
> 

Hans,
with 0707.4 tree I followed your steps to reproduce leaving the second disk for the second install. I got the error dialog telling me there wasn't enough free space. /so far bug is fixed/

After clicking OK there was a windows such as "re-scanning disks" or similar and then I just hit Next without changing previous settings. The second time anaconda didn't complain about missing space but happily removed my previous partitioning ( a single PV in this case) and created default partitioning (/boot, PV and / and swap in LVs). 

Is the behavior in the second retry expected or something related to this bug ?

--- Additional comment from hdegoede on 2010-07-14 20:59:12 EEST ---

Hi,

(In reply to comment #5)
> Hans,
> with 0707.4 tree I followed your steps to reproduce leaving the second disk for
> the second install. I got the error dialog telling me there wasn't enough free
> space. /so far bug is fixed/
> 
> After clicking OK there was a windows such as "re-scanning disks" or similar
> and then I just hit Next without changing previous settings. The second time
> anaconda didn't complain about missing space but happily removed my previous
> partitioning ( a single PV in this case) and created default partitioning
> (/boot, PV and / and swap in LVs). 
> 
> Is the behavior in the second retry expected or something related to this bug ?    

Sounds like you've found a new (related) bug, can you please reproduce and file a new bug for this with logs attached ?

Thanks,

Hans