HP-UX Logical Volume Manager and MirrorDisk/UX Release Notes HP-UX 11i v3 (March 2011)

Table 1 LVM Fixes in HP-UX 11i v3 March 2011 Release (continued)
DescriptionDefect ID
If automatic increase of pre-allocated extents is enabled for a space-efficient, distributed
snapshot, the pre-allocated extents are increased for each unsharing IO.
QXCR1001049913
1
During IOs on a source logical volume, the system panics. Reboot after panic generates the
following:
lvmp_cache_scrub
QXCR1001050822
1
If automatic increase of pre-allocated extents is enabled for a space-efficient snapshot, a
"WARNINGS"message is logged even after the size of snapshot reaches its original logical
volume.
QXCR1001050860
1
When trying to add a physical volume to a shared volume group in the MORE environment
using vgextend, the command fails because the pass-through controller and the data LUN
have the same wwid.
QXCR1001053066
1
In releases before HP-UX 11i v3 March 2011 and before cluster device special file (cDSF)
support was enabled, when an attempt is made to remove a logical volume (using lvremove)
in a shared volume group, LVM logs the following misleading messages in syslog; however,
lvremove successfully removes the logical volume:
lvmpud[3176]: Type mismatch /dev/lvtestvg/LVM21lv01
QXCR1001054379
The vgcfgbackup command may fail if there is inconsistency in the status of the device
path. This occurs when vgcfgbackup and pvchange are running in parallel, and the
physical volume has more than one physical links, and pvchange switches the active link to
another link.
QXCR1001057573
Each I/O in a distributed space-efficient snapshot causes automatic increase of pre-allocated
extents, if the distribution of free pre-allocated extents is located in only 2 physical volumes.
QXCR1001059016
When executing lvchange (with the -M y option) to set Mirror Write Cache ON for a
logical volume, the nomwcsyncd daemon process is not started and its usage string is
displayed on the console.
QXCR1001063396
A space-efficient snapshot is not marked as over-commit and inoperative and its data
may get corrupted if both the following conditions occur:
1. If allocation of pre-allocated extents of a space-efficient snapshot is not done for the first
time during creation of the snapshot or during activation of the volume group with the
snapshot
2. If a write IO is issued to its successor that requires data to be unshared on this space-efficient
snapshot and the IO on the successor completes.
This behavior is also seen in the case where the device special file (DSF) of the space-efficient
snapshot is deleted accidentally and subsequent activation fails to allocate the pre-allocated
extents to this snapshot.
QXCR1001063471
The vgextend command fails when trying to add a physical volume to a shared volume
group in the MORE HPVM environment.
QXCR1001063905
1
When a Version 2.2 volume group is activated in read-only mode, the snapshot tree is made
unavailable (for both the original logical volume and all the snapshots associated with it) if
there is a space-efficient snapshot in the tree.
QXCR1001064431
The vgchange command hangs indefinitely when quiescing the volume group which has
one or more mirrored snapshot logical volumes. This hang occurs only when the resync of
mirrored logical volume is triggered internally at the same time (for example, an offline physical
volume coming back online may trigger a resync).
QXCR1001064761
For a distributed space-efficient snapshot, automatic increase of pre-allocated extents is
activated way before the threshold is reached. This occurs when the physical volume's total
number of free pre-allocated extents (of all space-efficient snapshots associated with this
physical volume) which is mapped to this snapshot, is equal to or more than half of the number
of the snapshot's configured pre-allocated extents.
QXCR1001064952
6 Logical Volume Manager and MirrorDisk/UX Release Notes