Options

When you create a command specification, you can also change the options for executing the commands on the distributed hosts. Use the Options tab to view or change options.  You can change the number of hosts on which a command will run at the same time; verify that a host is responding before sending it commands, specify whether output will be streamed over time or appear all at once; and specify which remote shell to use.

You can find more detailed help on the following elements of this window:


Number of hosts to run commands on concurrently

The slider and text box both display the number of hosts on which the command specification will run at the same time.    Allowable values are between 1 and 64 hosts, with a default value of 64.  You can change this value by either dragging the slider to a new value, or by entering a value in the text box.  You may want to reduce the number of concurrent hosts if you are experiencing network problems and you want to improve performance.


Before running, verify that hosts will respond

Check this box if you want to make sure that hosts are online and responding before running a command specification.  By default, this checkbox is unchecked, meaning that commands will be sent to all hosts without checking whether they are available.  Changing this value means that you want to invest the time to check up front, which, if there are problems, should result in a smaller wait time than the minute typically taken for the remote shell command to time out.


Stream output

If checked, causes output to be streamed (displayed in the Execution Progress dialog as it is received).   If unchecked, causes output to be collected and displayed only after command execution completes.


Remote Shell

 Displays the remote shell under which distributed commands will be run. The remote shell options for rsh listed in this section are for the AIX platform.  On Linux, these options could be different. The default shell on AIX is  rsh, and if the secure remote shell is not installed, rsh will be the only option available.  You can enter options for either  rsh or the secure remote shell in the text box provided.  Enter the options as you would enter them on the command line.

Because the DCEM application is not interactive, you must configure the secure remote shell to run in batch mode.  If the secure remote shell is not configured properly, and you are prompted for a password during authentication, the command that you attempted to run cannot execute. You must then click the stop button at the bottom of the Execution Progress dialog to stop the execution.

Options for rsh include the following:

-f causes DCE credentials to be forwarded to remote hosts. This option is valid only if the underlying rsh uses Kerberos authentication and you have valid Kerberos credentials. It will be ignored if Kerberos 5 is not the current authentication method, and authentication will fail if the current DCE credentials are not marked forwardable.

-F causes the credentials on the remote system to be marked forwardable (allowing them to be passed to another remote system). This setting will also be ignored if Kerberos 5 is not the current authentication method.

DCEM supports a variety of secure remote shells and has been tested, to a limited basis, using OpenSSH, a secure remote shell.  If a secure remote shell supports batch mode, you must enable the batch mode. You can do this either in the secure remote shell client configuration file on the CSM server or in the secure remote shell options in the DCEM Options Tab. (If the secure remote shell is installed, you can select the Secure Remote Shell radio button, and in the field where the options for this selection display, you can type the flag that enables batch mode.)  For detailed information about secure remote shell client configuration, see the specific secure remote shell documentation.