Controlling mbatchd
Use the badmin reconfig, badmin mbdrestart, badmin mbdrestart -C, and bctrld stop sbd commands to control the mbatchd daemon.
Procedure
You
use the badmin command to control mbatchd.
Reconfiguring mbatchd
About this task
If you add a host to a host group, a host to a queue, or change resource configuration in the Hosts section of the lsf.cluster.cluster_name file, the change is not recognized by jobs that were submitted before you reconfigured.
If you want the new host to be recognized, you must restart the mbatchd daemon (or add the host that uses the bconf command if you are using live reconfiguration).
Procedure
Run the badmin reconfig command.
Results
When you reconfigure the cluster, mbatchd does not restart. Only configuration files are reloaded.
Restarting mbatchd
Procedure
Run the badmin mbdrestart command.
LSF checks configuration files for errors and prints the results to stderr. If no errors are found, LSF runs the following tasks:
- Reload configuration files
- Restart the mbatchd daemon
- Reread events in the lsb.events file and replay the events to recover the running state of the last instance of the mbatchd daemon.
Results
Tip: Whenever LSF restarts the mbatchd daemon, mbatchd is not available for service requests. In large clusters with many events in the lsb.events file, restarting the mbatchd daemon can take some time. To avoid replaying events in the lsb.events file, use the badmin reconfig command.