Enterprise Manager Server

Exporting data to XML from Control-M Enterprise Manager can be done from either the client or server. This example is based on extract from the client. To execute an xml report you need to have Control-M Client installed on the machine executing the command, the name of the scheduling table, the DC it's maintained and the following example text file, along with the command line. The text file should contain:

<?xml version="1.0" encoding="UTF-8"?> <?DOCTYPE TERMS SYSTEM "terms.dtd"> <TERMS> <TERM> <PARAM NAME="DATACENTER" OP="EQ" VALUE=""/> <PARAM NAME="TABLE" OP="EQ" VALUE=""/> </TERM> </TERMS>

And this is the command line that executes against this text file.

C:\>ExportDefTable -HOST <EM-hostname> -USERNAME <user> -PASSWORD <password> -ARG_FILE c:\<path\file.txt> -OUT_FILE c:\<path\file.xml>

Control-M Enterprise Manager (EM) GUI is slow to respond, sysout not displaying or slow to display. This may actually be either a Control-M Server issue, a Gateway issue or memory leak. If the gateway appears to be functioning correctly, click here to investigate the Control-M Server.

The memory leak in 6.2.01 was fixed in FixPack5. Prior to FP5 the corrective action usually resulted in rebooting the EM GUI Server. On Unix you can run nmon after a reboot to see the memory usage and check back in a few weeks to see if it continues to build or stays static.

For a Gateway issue, it may be a Control-M server issue, click here to review steps. If not, try to bounce the gateway and/or rebuild it. If this doesn't resolve, contact BMC support.

Delete from AJF, hung batch, stays in yellow status, white job. First thing that must be done is confirmation the job is no longer needed in the EM/GUI. Did the job finish? Is the Agent stable, does it needed to be cleared and restarted? If there is no doubt the job is ready to be removed from the AJF, one of the easiest ways to accomplish this task is to rename the job "remove", run ctm_menu, Option 10 Troubleshooting, Option 14 for db access and run (on Sybase):

delete from CMR_AJF where JOBNAME='remove' go

That will delete from the AJF all the jobs that are named "remove". At this point the jobs will remain until new day processing executes.

DISAPPEARED job... A job can become "disappeared" from three different scenarios.

1. If a Control-M/Agent becomes unknown to the Control-M/Server due to loss of communication and stays unknown for a period of about two hours, then all the jobs that were sent to this Control-M/Agent, which are in status "submitted", or "executing", will be converted, by the Control-M/Server to disappeared. So the decision on "disappearing" a job is initiated by the Control-M/Server.

2. A job has been submitted to and executed by the Control-M/Agent, but for some reason the application that the job runs has been terminated abnormally. In this case, the Control-M/Agent does not detect the crashing of the application. The Control-M/Server still thinks the job is executing. When the Control-M/Server sends the track-all message to the Control-M/Agent asking about this particular job, the Control-M/Agent will realize that the application the job runs cannot be found, and it will tell the Control-M/Server to change the job's state to disappeared.

3. The Control-M/Agent has crashed or has been shutdown by an administrator. In this case, all the jobs that were in a "submitted" or "executing" state on the Control-M/Server will become "disappeared", when the agent is brought up. Simply wait for the Control-M/Server to initiate a check to the Control-M/Agent after the problem causing the particular issue has been resolved and the job status will be updated. In the case of a failed job, rerun the job after resolution has been applied. Keep in mind, some cleanup might have to be done to resolve any jobs which are not automatically handled. In this case, hold the job and delete it (for delete instructions, click here). Then, order the particular job back in to the AJF and allow it to run and complete.

FORKED job... The forked job is an indication that two ctmagent processes have been started. Shutdown the agent, clear (kill) any open processes and restart the agent.

GUI on the laptop does not display. Occasionally there have been reports of laptop users having problems accessing the GUI and displaying the view point. One of the reasons for this appears to be related to the network default connection and/or a setting. To ensure the correct setup open your network connection page, on the tool bar under Advance click on Advance Settings and adjust your priority connections accordingly.

Typically the wireless connection is first. If you are hardwired, move the wireless connection down and your LAN connection up. If you are running wireless (normally I don't see this, but) move the wireless up if there are problems. the wireless is usually not an issue so keeping it at a lower priority should be sufficient.

Settings for the laptop funtioning correctly, follow these steps: 1. On the GUI Client machine (typically a laptop) goto a command prompt, execute the following command: orbadmin variable create -scope default -value iiop:///hostname_in_ior=yourServerHere -ORBListenEndpoints 2. The output should be: "Variable '-ORBListenEndpoints' added/updated successfully" 3. This will add a line to the config.xml file in the default scope section that looks like: <variable name="-ORBListenEndpoints" value="iiop:///hostname_in_ior=gsrhostname"/>