Serviceguard NFS Toolkit A.11.11.06, A.11.23.05 and A.11.31.03 Administrator's Guide

of a node or network failure. The NFS monitor script causes the package failover if any of
the monitored NFS services fails.
If you do not want to run the NFS monitor script, comment out the NFS_SERVICE_NAME
and NFS_SERVICE_CMD variables:
# NFS_SERVICE_NAME[0]=nfs.monitor
# NFS_SERVICE_CMD[0]=/etc/cmcluster/nfs/nfs.mon
By default, the NFS_SERVICE_NAME and NFS_SERVICE_CMD variables are commented
out, and the NFS monitor script is not run.
NOTE: The Serviceguard A.11.13 configuration file includes the following changes:
• AUTO_RUN replaces obsolete PKG_SWITCHING_ENABLED.
• LOCAL_LAN_FAILOVER_ALLOWED replaces obsolete NET_SWITCHING_ENABLED.
4. To enable File Lock Migration (available on 11i v1 and 11i v2), set the
NFS_FILE_LOCK_MIGRATION variable to 1:
NFS_FILE_LOCK_MIGRATION=1
By default, this variable is set to 0 (disabled). The NFS_FLM_SCRIPT variable is the name
of the script that manages synchronization of the file lock status entries for the primary and
adoptive nodes associated with this HA/NFS package. By default, this is set to nfs.flm.
You must assign a unique name to this script in every HA/NFS package in the cluster (for
example, nfs1.flm, nfs2.flm, and so on):
NFS_FLM_SCRIPT="${0%/*}/nfs1.flm"
If you wish to monitor the File Lock Migration script, then you must also set the
NFS_FILE_LOCK_MIGRATION and NFS_FLM_SCRIPT variables in the NFS monitor script.
If you enable File Lock Migration, then you can configure the File Lock Migration script (see
“Editing the File Lock Migration Script (nfs.flm)” (page 31)).
NOTE: The file name of the NFS_FLM_SCRIPT script must be limited to 13 characters or fewer.
Editing the File Lock Migration Script (nfs.flm)
The File Lock Migration script, nfs.flm, handles the majority of the work involved in maintaining
file lock integrity that follows an HA/NFS failover. The nfs.flm script includes the following
configurable parameters:
• NFS_FLM_HOLDING_DIR - Name of a unique directory created in one of the shared volumes
associated with this package. This directory holds copies of the /var/nfs4/v4_state
files for this package. You must create this directory in one of the shared volumes associated
with this package so that it can migrate with the package (from the primary server to the
adoptive server).
You must dedicate this directory for holding v4_state entries only. In addition, you must
keep it empty. This directory should not have other files or subdirectories when starting the
cluster. All files in this directory are deleted after a failover. . This directory must be located
in the same shared volume as NFS_FLM_HOLDING_DIR.
An example for this parameter is as follows:
NFSv4_FLM_HOLDING_DIR="/pkg1a/v4_state"
• PROPAGATE_INTERVAL - Number of seconds between the attempts of the script to copy
files from the /var/statmon/sm directory into the holding directory, specified by
NFS_FLM_HOLDING_DIR. The default value of this parameter is five seconds.
An example for this parameter is as follows:
PROPAGATE_INTERVAL=5
Configuring a Serviceguard NFS Package 31