Managing Serviceguard 13th Edition, February 2007

Troubleshooting Your Cluster
Solving Problems
Chapter 8 371
This can happen if a package is running on a node which then fails before
the package control script can deport the disk group. In these cases, the
host name of the node that had failed is still written on the disk group
header.
When the package starts up on another node in the cluster, a series of
messages is printed in the package log file, as in the following example
(the hostname of the failed system is ftsys9, and the disk group is dg_01):
check_dg: Error dg_01 may still be imported on ftsys9
To correct this situation, logon to ftsys9 and
execute the following command:
vxdg deport dg_01
Once dg_01 has been deported from ftsys9,
this package may be restarted via either cmmodpkg(1M)
or cmrunpkg(1M).
In the event that ftsys9 is either powered off
or unable to boot, then dg_01 must be force
imported.
******************* WARNING**************************
The use of force import can lead to data corruption if
ftsys9 is still running and has dg_01
imported. It is imperative to positively determine that
ftsys9 is not running prior to performing the force
import. See -C option on vxdg(1M).
*******************************************************
To force import dg_01, execute the following
commands on the local system:
vxdg -tfC import $vg
vxdg deport $vg
Follow the instructions in the message to use the force import option (-C)
to allow the current node to import the disk group. Then deport the disk
group, after which it can be used again by the package. Example:
# vxdg -tfC import dg_01
# vxdg deport dg_01