HP-UX Reference (11i v3 07/02) - 1M System Administration Commands N-Z (vol 4)

v
vgchange(1M) vgchange(1M)
NAME
vgchange - set LVM volume group availability
SYNOPSIS
Activate volume group
/usr/sbin/vgchange -a
availability [-l][-p
][-q quorum][-s][-P resync_daemon_count]
[vg_name ...]
Set volume group high-availability cluster attributes
/usr/sbin/vgchange -c
cluster [-S sharable] vg_name
Change activation mode of sharable volume group
/usr/sbin/vgchange -a
availability -x {vg_name...}
Quiesce an active volume group
/usr/sbin/vgchange -Q
quiesce_mode [-t quiesce_time] vg_name
Resume a quiesced volume group
/usr/sbin/vgchange -R
vg_name
Remarks
MC/ServiceGuard cluster operations require the installation of the optional MC/ServiceGuard software,
which is not included in the standard HP-UX operating system.
Mirrored disk operations require the installation of the optional HP MirrorDisk/UX software, which is not
included in the standard HP-UX operating system.
DESCRIPTION
The
vgchange command with the -a option activates or deactivates one or more volume groups.
For volume groups that are used in a high availability cluster, when the -a
option is specified together with
the
-x option, the vgchange command allows cross-activation, changing the activation mode of an active
sharable volume group directly to the specified activation mode, without first having to deactivate the
volume group.
The vgchange command with the -c specifies whether the indicated volume group(s) can be used in a
high availability cluster, optionally the -S option can also be specified to configure whether the volume
group is sharable (or is for exclusive activation only).
The vgchange command without the -P resync_daemon_count option (default) will spawn one
nomwcsyncd process for each volume group containing
NOMWC logical volumes, and this can degrade sys-
tem performance when many of these volume groups are activated at the same time. The -P
resync_daemon_count option provides a way to control the number of concurrent nomwcsyncd processes
spawned. This parameter is advisory, LVM may use a different value internally as necessary.
The
-Q option allows quiescing an active volume group to facilitate creating a consistent disk snapshot copy
of all the physical volumes in the volume group. Metadata on the physical volumes is brought current and
remains unchanged until normal operation is resumed using the -R option (or when the optionally specified
-t quiesce_time time has expired). Specifying -Q rw when quiescing the volume group quiesces applica-
tion reads and writes to the volume group. This mode is suitable for most situations. Specifying -Q w
when quiescing the volume group quiesces application writes to the volume group while allowing reads to
proceed as usual. This mode provides improved access to the data when used with applications that are
capable of quiescing updates while allowing queries. See the Quiesced Volume Groups section for a more
complete description of quiesced volume groups.
vg_name must be defined as a volume group in the file /etc/lvmtab .Ifvg_name is omitted, all volume
groups defined in /etc/lvmtab are affected, except when the -c -x -Q or -R options are specified.
Only a single volume group can be provided with these options.
High Availability Cluster Overview
Volume groups can be defined on disk volumes that are accessible by more than one system in a high avai-
lability cluster. This situation has a high potential for data corruption unless higher level cluster software
services are used to coordinate shared access to the same volume group by all systems.
A volume group can be marked as part of a cluster. When such a group is activated in exclusive mode, it
can be accessed for exclusive read-write activity by only one of the systems at a time; the other systems can
618 Hewlett-Packard Company − 1 − HP-UX 11i Version 3: February 2007