VxVM disk groups (other than those managed by CVM) are outside
the control of the Serviceguard cluster. The package control script
uses standard VxVM commands to import and deport these disk groups.
(For details on importing and deporting disk groups, refer to the
discussion of the import and deport options in the vxdg man page.)
The control script imports disk groups using the vxdg command with the -tfC options. The -t option specifies that the disk is imported with the noautoimport flag, which means that the disk will not be automatically re-imported
at boot time. Since disk groups included in the package control
script are only imported by Serviceguard packages, they should not
be auto-imported.
The -f option allows the disk group to be imported even if one
or more disks (a mirror, for example) is not currently available.
The -C option clears any existing host ID that might be written
on the disk from a prior activation by another node in the cluster.
If the disk had been in use on another node which has gone down
with a TOC, then its host ID may still be written on the disk, and
this needs to be cleared so the new node’s ID can be written
to the disk. Note that the disk groups are not imported clearing
the host ID if the host ID is set and matches a node that is not
in a failed state. This is to prevent accidental importation of
a disk group on multiple nodes which could result in data corruption.
 |
| |
|
 | CAUTION: Although Serviceguard uses the -C option
within the package control script framework, this option should
not normally be used from the command line. Chapter 8 “Troubleshooting
Your Cluster”, shows some situations where you might
need to use -C from the command line. |
|
| |
|
The following example shows the command with the same options
that are used by the control script:
# vxdg -tfC import dg_01
This command takes over ownership of all the disks in disk
group dg_01, even though the disk currently has a different host
ID written on it. The command writes the current node’s
host ID on all disks in disk group dg_01 and sets the noautoimport flag
for the disks. This flag prevents a disk group from being automatically
re-imported by a node following a reboot. If a node in the cluster
fails, the host ID is still written on each disk in the disk group.
However, if the node is part of a Serviceguard cluster then on reboot
the host ID will be cleared by the owning node from all disks which
have the noautoimport flag set, even if the disk group is not under
Serviceguard control. This allows all cluster nodes, which have
access to the disk group, to be able to import the disks as part of
cluster operation.
The control script also uses the vxvol startall command to start up the logical volumes in each disk
group that is imported.
Printable version
