10.4.1 Script for forcibly stopping JP1/AJS3 (jajs_killall.cluster)
This subsection gives an overview of the shell script for forcibly stopping JP1/AJS3 (jajs_killall.cluster).
For details about the syntax of the jajs_killall.cluster shell script for forcibly stopping JP1/AJS3, see jajs_killall.cluster (UNIX only) in 3. Commands Used for Normal Operations in the manual JP1/Automatic Job Management System 3 Command Reference.
The following table lists the platforms and products supported by the shell script for forcibly stopping JP1/AJS3.
HP-UX |
Solaris |
AIX |
Linux |
|
---|---|---|---|---|
JP1/AJS3 - Manager |
Y |
Y |
Y |
Y |
JP1/AJS3 - Agent |
Y |
Y |
Y |
Y |
- Legend:
-
Y: Supported
- Cautionary notes
-
-
Use the jajs_killall.cluster command only when the process does not terminate after the service is terminated by the jajs_stop.cluster command.
-
To try to restart JP1/AJS3 on the same node, after stopping JP1/AJS3 by executing jajs_stop.cluster, execute a forced stop (ajs_killall.cluster command) for clean-up and then start JP1/AJS3 by executing jajs_start.cluster command.
-
The shell script for forcibly stopping JP1/AJS3 stops the processes for logical hosts. However, this shell script does not terminate the following JP1/AJS3 processes:
- jpocwtmqmaii
MQ monitoring process when JP1/AJS3 is linked with MQSeries.
You do not need to fail over the above processes together with the JP1/AJS3 service when a cluster system is used. Stop them as required.
-
- Supplementary note
-
This command outputs a log entry to a file on the shared disk. If the shared disk cannot be accessed, the log cannot be output and therefore JP1/AJS3 cannot be forcibly stopped. In this case, change the output destination for logs to a directory on the physical disk.
To change the output destination for logs to a directory on the physical disk:
-
Use an editor such as vi to open the following file (the shell script for forcibly stopping JP1/AJS3):
/etc/opt/jp1ajs2/jajs_killall.cluster
-
Change the file as follows:
-