On the host where vmd is the recipient of the connection, verify that the daemon or the service is running. (The host is the Media Manager host, the Device Host, or the EMM server.) If the daemon or the service is not running, start it. On Windows, vmd is the NetBackup Volume Manager service.

If vmd is already running, examine the command output, debug logs, and system logs for a more detailed message on the error.

See "Setting Media Manager debug logging to a higher level" in the NetBackup Logging Reference Guide.

Verify that the correct host names are defined in the configuration.

Check the services file. On UNIX and Linux, verify that the /etc/services file (and NIS services if NIS is used) has entries for the vmd service. (Note that the vmd service always starts oprd.) On Windows, verify that the %systemroot%\system32\drivers\etc\services file has the correct entry for vmd. Also verify that the vmd port number in the services file agrees with the port number configuration. The port number is noted in the man page for vmd(1M).

Verify that all of the operating system patches or service packs are installed.

Ensure that the Media Manager configuration is not tuned so that the load on vmd exceeds its ability to service requests. Look for entries in the vm.conf file that increase the load. Consider placing the EMM database on a higher performance server and file system if performance is an issue. To reduce the number of volumes in the volume configuration, use inventory filtering for the robot types that support it

Check the utilities, such as ipcs -a, to ensure that the shared memory functions properly. The oprd process may not respond because it cannot attach to shared memory.