jobrecovery.bat|.sh batch script
We can use the jobrecovery script to enable a secondary site to take over when the primary site fails.
Purpose
In this scenario, each site has a separate WebSphere cell where an active job is running on the primary site, and a stopped job or job scheduler is on the secondary site. The batch cells on the primary and secondary sites are expected to be clones of one another with the same topography and the same names. They share the same database, not replicas of a database.
Location
The jobrecovery script is located either in the app_server_root/bin directory or in the app_server_root/profiles/{LRS_profiles}/bin directory on the designated LRS backup server. If the script is run from the app_server_root/bin directory, the profileName option must be specified. Otherwise, do not specify the profileName option. The user name and password options are required if security is enabled on the server which is running the LRS. Run on the backup cell after the primary cell is shut down. After it is run, the backup cell can be started.
Usage
When the primary site fails, you are instructed to follow a takeover procedure that includes the following steps:
- Ensure that all batch servers on the primary site are stopped.
- Run the jobrecovery script on the secondary site.
- Start the job scheduler on the secondary site.
./jobRecovery.sh server [options]
Options
- -profileName profile
- Use this optional parameter to specify the stand-alone application server profile name when the script is run from the app_server_root/bin directory.
- -username authentication username
- Use this optional parameter to specify a user name.
- -password authentication password
- Use this optional parameter to specify an authentication password for the user name.
Example
./jobRecovery.sh server1 -profileName AppSrv01 -username wsadmin -password wspassword
Related concepts
Configure the batch environment
Parallel job manager APIs