5.8.3 Notes on cluster use (limited to UNIX only)
-
The following procedure shows how to change the language setting for a logical host after setting up the cluster system environment. Shut down JP1/Base and related programs before proceeding.
-
Change the language setting in the shared file jp1bs_env.conf on the shared disk used by the logical host whose language you want to change.
For details on how to do so, see 3.4.2 Setting the language (for UNIX).
-
Create a text file in vi or another editor.
In this example, the language is set to EUC (Japanese) encoding.
[logical-host-name\JP1BASE\]
"LANG"="EUCJIS"
End the final line with a linefeed character.
-
Save the text file you created.
You can give the file any name. In this example, the file name is baselang.conf.
-
Execute the following command as a user with superuser or JP1/Base administrator permission.
/opt/jp1base/bin/jbssetcnf baselang.conf
-
Edit the automatic start script for logical hosts (jbs_start.cluster).
Set the LANG environment variable of the automatic startup script for logical hosts (jbs_start.cluster) to the same language specified in jp1bs_env.conf in step 1.
If you manually start a logical host event service without using the automatic start script (jbs_start.cluster), locale information used when the event service is started (for example, the LANG environment variable) must match the language specified in jp1bs_env.conf.
If, for example, you specify ja_JP.UTF-8 as the language for jp1bs_env.conf, change the LANG environment variable of the automatic startup script for logical hosts (jbs_start.cluster) as follows:
## Set Environment-variables
PATH=/sbin:/bin:/usr/bin:/opt/jp1base/bin
LANG=ja_JP.UTF-8
SHLIB_PATH=/opt/jp1base/lib:/opt/hitachi/common/lib
-
-
If you stop JP1/Base services that support logical hosts, a JP1/Base process might fail to stop. In this case, execute the jbs_killall.cluster command to forcibly stop the process. Note that this command should be used only when JP1/Base processes cannot be stopped with the normal procedure. For details on the jbs_killall.cluster command, see jbs_killall.cluster (UNIX only) in 15. Commands.
-
In a cluster system that performs monitoring even during stop processing, modify the command that terminates JP1/Base (the event service and process management including user management), as follows:
cd /etc/opt/jp1base cp -p jbs_stop.cluster.retry.model jbs_stop.cluster
-
If a shell script for monitoring the operating status is registered in the cluster system, and the monitoring interval of the jbs_spmd_status command used by the shell script is a short time of 10 seconds or so, the next jbs_spmd_status command may execute before the previous jbs_spmd_status command completes, potentially causing an unexpected failover. The monitoring interval should therefore be at least 60 seconds. Also, if you changed the jbs_spmd_status command's timeout (-t option) from the default, set the monitoring interval to a longer time than the time set in the -t option.