HP-UX Logical Volume Manager and MirrorDisk/UX Release Notes (March 2010 Update release)

For more information, see lvm(7) and lvmove(1M).
Bootable Version 2.2 and higher volume groups
LVM now enables you to create bootable Version 2.2 and higher volume groups. LVM always
provided the ability to create bootable Version 1.0 volume groups in the HP-UX 11i v1, 11i v2,
and 11i v3 releases. Despite this new support, you cannot create bootable Version 2.0 and 2.1
volume groups.
For more information, see lvlnboot(1M), lvm(7), and vgcreate(1M).
Changed features
There are no changed features in the March 2010 release of LVM and MirrorDisk/UX.
Deprecated or obsolete features
No LVM or MirrorDisk/UX features are deprecated or obsolete in the March 2010 release of
HP-UX 11i v3.
Known problems fixed in this version
The following table lists the known LVM and MirrorDisk/UX problems fixed in the March 2010
release of HP-UX 11i v3.
Table 1 LVM Fixes in HP-UX 11i v3 March 2010
DescriptionDefect ID
During a fresh installation or in cases where the /etc/lvmconf/pv_lock file has been
deleted, the following message is logged in the syslog file: /etc/lvmconf/pv_lock
file was found to be of zero size. There could be delay or failure
in the generation of unique PVIDs. This informational message has been reworded
to avoid confusion.
QXCR1000936867
In low memory conditions, the pvmove command can fail on Version 2.x volume groups
and cause subsequent Version 2.x LVM commands to hang.
QXCR1000940488
If a physical volume in a Version 2.x volume group is restored using the vgcfgrestore
command and the physical volume contains mirrored logical volumes with Mirror Write
Cache enabled (as shown by lvdisplay), user I/O to those logical volumes might
subsequently fail with EIO.
QXCR1000940892
In some instances, the pvmove command fails with the following error: pvmove: Couldn't
retrieve the allocation map of physical volume.
QXCR1000945735
For Version 2.0 and higher volume groups, if the volume group has more than 32 physical
volumes, a pvmove auto rebalance does not happen across all the physical volumes properly.
QXCR1000954120
If you use the Version 2.x volume group resizing feature (-S option) of the vgmodify
command to increase or decrease the maximum size of a 2.x volume group and then use
the vgextend command to add physical volumes to the volume group, the volume group
metadata can enter an inconsistent state such that subsequent activations of the volume
group can fail with a message similar to the following: Quorum not present, or some
physical volume(s) are missing. In addition, commands that attempt to modify
the volume group configuration can fail.
QXCR1000967236
Booting a volume group using maintenance mode boot can result in the corruption of the
root file system. The symptom is an error message to the system console on the next boot
similar to the following: fsadm: ERROR: V-3-20295: device record checksum
is invalid.
QXCR1000969018
For Version 2.0 and higher volume groups, pvmove dumps core when a logical volume is
moved to the primary and alternate link of a physical volume.
QXCR1000974930
For Version 2.0 and higher volume groups, vgextend does not always succeed for an
alternate link of a physical volume that is already a part of other volume group.
QXCR1000975395
6