HP-UX Logical Volume Manager and MirrorDisk/UX Release Notes HP-UX 11i v3 (October 2010)

Known problems
Auto pre-allocation occurs for all unsharing if distributed snapshot has 2 PVs
Defect ID: QXCR1001059016
Problem: The default threshold for a distributed, space-efficient snapshot is
reached when the number of free extents in the pre-allocated pool
is less than or equal to 20 percent of the total number of extents in
the pre-allocated pool or when more than half of the number of free
extents in the pre-allocated pool is from the same physical volume.
If the distribution of the snapshot has only 2 physical volumes, at
any point in time, a single physical volume always has 50% or less
free pre-allocated extents. So, auto pre-allocation will always occur.
Severity: Medium
Corrective Action: Configure the data distribution such that it lies on more than
2 physical volumes.
•
• As auto-preallocation occurs for each unsharing IO, it is possible
that the size of the snapshot will reach the original logical
volume size. Ensure that there are enough free extents in the
volume group.
Auto pre-allocated, distributed snapshot threshold considers all LV pre-allocated extents in PV
Defect ID: QXCR1001064952
Problem: While calculating the number of pre-allocated extents in a physical
volume for a distributed space-efficient snapshot for getting the
threshold value, LVM needs to consider the pre-allocated extents
belonging to the particular logical valume. But, instead, it considers
all logical volumes' pre-allocated extents.
Severity: Medium
Corrective Action: Configure your system such that one physical volume does not
have more than one space-efficient snapshot.
•
• As auto-preallocation occurs for each unsharing IO, it is possible
that the size of the snapshot will reach the original logical
volume size. Ensure that there are enough free extents in the
volume group.
Activation of Version 2.2 VG with space-efficient snapshots fails to properly mark snapshots as
read-only when there are no free extents
Defect ID: QXCR1001071538
Problem: When the activation of a version 2.2 volume group with
space-efficient snapshots fails to allocate pre-allocated extents due
to lack of free extents, the vgchange command should mark the
snapshots as read-only to prevent the snapshots from becoming
over-committed and inoperative. Instead, the vgchange command
displays “Invalid argument” and fails to mark the
space-efficient snapshot logical volumes as read-only. This behavior
is only seen for space-efficient snapshot logical volumes with
read/write permissions.
Severity: Serious
Known problems and limitations 7