Enterprise User Manual

Chapter 5

Working with Operations

---

Operations are actions performed from a PC-Duo Enterprise Console on selected Clients or Groups. Examples operations include Hardware Inventory, Software Inventory and Software Distribution.

Operations are displayed in the Console's Operations View at the bottom part of a PC-Duo Enterprise Site window. Related operations are organized into various tabs.

Performing Operations

To perform an operation, select the Group or Nodes on which you want to perform the operation, then double-click the operation's icon.

Initially, some operation icons feature a red cross. This indicates that these operations must be customized before they can be used.

Operations and Clients

When working with standard PC-Duo Enterprise Clients, operations run against the existing Client data in the Offline Area. The frequency which the Client data is pushed to the Offline Area is specified at the Client. As such, the offline Clients themselves do not have to be switched on or accessible over the network to retrieve Client data.

Customizing Operations

PC-Duo Enterprise offers great flexibility by enabling you to create new, customized variants of operations, and to also configure existing operations.

Customizing operations into variants means you can match the way you manage your PCs to the particular needs of each group of users and the needs of your PC management policies. For example, you may decide that when performing an inventory of a particular department, you want to scan for unrecognized applications. In such a case, you can create a Software Inventory (Unrecognized) Operation variant that contains the settings you want to use.

To Create a New Operation Variant:

1. Right-click in the Console's Operations View, then choose the New command from the shortcut menu.
2. In the Add New Operation dialog, select the type of operation you want to create from the Type drop-down list. The selected type determines the options you can choose for the new operation and the tabs in which it will appear.
3. Click the OK button. A Customize dialog is displayed (the name of this dialog depends upon the operation you are customizing).
4. Enter an Operation name. The name must be unique.
5. Update the settings, then click the OK button. The new operation is displayed.

   If you can't see the new operation, make sure you have selected an appropriate tab in the Operations View, or select the All tab, which lists all operations available at the Site.

To customize an existing Operation:

1. Right-click on the operation you want to customize, then choose the Customize command from the shortcut menu.
2. In the customize dialog, configure the options in the dialog tabs, then click the OK button to save the customized operation.

   You can now use this customized operation on the Nodes and Groups at this Site.

Deleting Operations

If you decide you no longer need an operation at a Site, you can delete it from the Operations View by right-clicking the operation and choosing Delete from the shortcut menu.

If you delete an operation, it is deleted from the Database. This means that any Consoles or Schedulers sharing that Site Database will not be able to use it.

If you delete all the operations of a particular type, you can create new operations of that type using the New command in the shortcut menu of the Operation View.

Importing and Exporting Operations

You can import and export operations using the Import Operations and Export Operations commands in the Operation View's shortcut menu.

Operation details are stored in .OBF files (Operation Backup Files), which can be transferred from one PC to another.

The .OBF files is stored in your PC-Duo Enterprise installation folder, and will be read by the Console application when it next loads.

Scan on Demand

Clients which are fundamentally Offline can be scanned interactively using the Scan On Demand mode. The Nodes being scanned in this fashion must be switched on and accessible over the network.

Publishing software inventory data from a Scan On Demand Operation is also possible through an alignment facility, which enables you to specify how the share names map onto physical drives on the Client PCs.

When you perform an operation in Scan on Demand mode, PC-Duo Enterprise analyzes the shares available on the relevant Clients and displays a list of shares, enabling you to select the ones to be processed.

Activating Scan on Demand

The Scan on Demand button in the Console toolbar switches Scan on Demand mode on and off. To activate Scan on Demand, click on the Scan on Demand button in the toolbar. Operations which support Scan on Demand run in this mode until the button is clicked again.

Configuring Scan on Demand Mode

You can configure the Scan on Demand facility to use a default connection profile, which PC-Duo Enterprise uses to connect to Client PCs when working in Scan on Demand mode. This default connection profile consists of a user name and password combination.

These settings are specified using the Scan on Demand Options dialog. This dialog is also used to specify whether you want to connect to PCs using their Admin shares or their user shares.

To configure Scan on Demand mode:

1. At the Console, choose the Site, Scan on Demand Options menu command. The Scan on Demand Options dialog is displayed.
2. Check the Use default connection profile box to specify a default user name and password for Scan on Demand to use when attempting to connect to Clients.
3. If necessary, enter and verify the User name and Password to be used as the default connection profile.
4. Select the Use admin shares or Use user shares option, then click OK to save these settings.

Configuring Clients to use Scan on Demand

Scan on Demand uses shares on Client PCs. For Windows NT4, 2000 and XP Clients, these shares are typically available as standard Admin Shares.

On Windows 95, 98 and Me Client systems, you may need to enable Scan on Demand using one of the following methods:

To edit the LUCLIENT.INI file for Scan on Demand:

1. On the Client PC, locate the file LUCLIENT.INI. This file is located in the \PC-Duo Enterprise directory by default.
2. Open this file using Notepad, and change the line beginning EnableFileSharing= to:
   EnableFileSharing=Yes
3. Locate the line beginning SharedDrives= and change the default setting of C,D,E,F to the drives that you want to share on the Client PC. For example:
   SharedDrives=C,D
4. Save the changes to the LUCLIENT.INI file. The changes take effect when the Client next checks the Offline Area.

To prepare Clients for Scan on Demand mode:

1. At the Console application, right-click on the Client Configuration Operation and choose Customize from the shortcut menu.
2. When the Customize Client Configuration dialog appears, select the Client Configuration tab. This tab contains a range of Client settings which you can reconfigure.
3. In the Update sections box, select the Boot control check box. The Boot control section on the right of the dialog can be configured.
4. In the Boot control section, select the Enable file sharing check box. In the Shared drives box, enter a comma-separated list of the drives you want to share - for example C,D,E.
5. Leave the rest of the settings untouched, then click OK to save the settings for the operation and return to the Console.
6. In the Console's Site window, select the Clients (or the Group) that you need to customize in order to enable Scan on Demand.
7. Double-click the Client Configuration Operation. The Configuration changes take effect on the next Client startup.

Publishing Scan on Demand Results

Publishing inventory data from a Scan On Demand Operation is possible through an alignment facility, which enables you to specify how the share names map onto physical drives on the Client PCs.

If you publish the results of an inventory operation performed using Scan on Demand mode, the Align Shares... dialog is displayed, enabling you to associate Scan on Demand shares with the stored shares in the Database.

To use the Align Share dialog:

1. Select a share from the Scan on demand shares box, then select the share from the Stored shares box which you want to associate with it.
2. Use the Link button to associate the selected shares. The shares are displayed in the Share links box.
3. Use the Remove button if you want to delete the association between two shares in the Share links box.
4. Click the OK button to publish the Scan on Demand results into the Database.

Scheduling Operations as Jobs

You can schedule operations by submitting them as Jobs to be performed by a PC-Duo Enterprise Scheduler, at a specified date and time.

A Job is an association of an operation, with a target selection of Groups and Clients, plus scheduling and logging information.

Job Monitoring Capability

After submitting a Job, you can monitor its progress using the Jobs Database window, which shows the Jobs currently in Progress. As soon as a Job has completed, it is transferred to the Job Log Database window.

You can view these windows using the Show Jobs button and Shows Job Logs button in the Console toolbar.

Jobs and Schedulers

Jobs are executed by PC-Duo Enterprise Schedulers. On Windows 9x and Me PCs, this is the WinScheduler application. On Windows NT4, 2000, and XP, you can use the Scheduler Service.

Operations You Can Schedule

In PC-Duo Enterprise, you can schedule the following operations:

Auto-Populate

Populates your Site with new Clients.

Hardware Inventory

Collects hardware asset information from Clients, such as network shares, drive sizes, and video settings.

Software Inventory

Lists the applications and components installed on the Clients.

User Prompt Refresh

Presents Client PC users with questions and retrieves answers. It can also collect environment variables and INI file settings.

Software Distribution

Deploys and silently installs applications on Clients.

Command Launch Operations

Run external operations, such as the Database Viewer, which have been added to your Operations View.

Working with Schedulers

Schedulers manage and perform Jobs which have been submitted using a PC-Duo Enterprise Console on a single Site Database. They run using the account and password that is specified during the Console installation.

You can use the PC-Duo Enterprise WinScheduler application to manage scheduled Jobs. This is a standard Windows application which is started manually from the Start menu or from the PC-Duo Enterprise Console.

If you want to use a Scheduler on a Windows 95, 98 or Me system, use the WinScheduler application. On Windows NT4, 2000 and XP, you can run the Scheduler service. Running the Scheduler as a Service is preferable because services are not dependent on users being logged on.

It is not possible to run the two types of Scheduler on the same PC simultaneously - the WinScheduler application always checks that the Service is running.

The Scheduler service is therefore able to run in the context specified in the Windows Control Panel's Services dialog. This enables it to automatically run, without requiring anyone to log on.

Although the Scheduler service supports the "AT" command, it is independent of the "AT" Scheduler. The PC-Duo Enterprise WinScheduler application only supports PC-Duo Enterprise Jobs

Installing Schedulers

The PC-Duo Enterprise Scheduler is installed with the Console. As Schedulers access the Site Database directly, after PC-Duo Enterprise is installed you can use the PC as a Scheduler Host.

Initially, the Console is used to configure a Host's access to a Site Database using the Site Creation Wizard. You can then use the Console and its WinScheduler Operation to reconfigure the Scheduler.

Starting Schedulers

You can either use PC-Duo Enterprise's WinScheduler application as your Scheduler. On Windows NT4, 2000 and XP systems, you can use the Scheduler Service.

Starting the Scheduler Service

When you install a Scheduler Service on Windows NT4, 2000 or XP, you can start it manually or automatically.

If you selected a Manual start for the Scheduler, you can configure it for automatic startup using the Windows Control Panel's Services dialog.

Starting the WinScheduler Application

On Windows 9x and Me, the Scheduler runs as a Windows application called WinScheduler.

This application typically resides in the taskbar's system tray. To start it, double-click the WinScheduler Operation in the PC-Duo Enterprise Console, or use the PC-Duo Enterprise Windows Scheduler command in the Start menu's PC-Duo Enterprise program folder.

To stop the WinScheduler, double-click on its icon to display the WinScheduler Properties dialog, then click the Stop button.

Checking Schedulers from the Console

Use the Console's Jobs Database window to identify operations that can be run by a Scheduler. To display this window, click the Show Jobs button in the Console toolbar.

The bottom half of this window, called the Hosts View, shows the status of the PC-Duo Enterprise Schedulers:

• A Host icon with no colored light means the Host is offline.
• A Host icon with a green light means the Host is online.
• A Host icon with a red light means the Host is shutting down.

You can view the Host Details for a Scheduler by right-clicking on a Host icon in the Jobs Database window and choosing the Properties command from the shortcut menu.

If a Host definition has just been added through the Console, but the Scheduler has not yet started, the definition indicates that it can support all operation types.

When the Scheduler starts, it defines its own Job capabilities by writing the correct list of supported operations to the Database.

Configuring Schedulers

Several Scheduler properties can be changed to tune the way that a Scheduler processes Jobs:

Wakeup Interval

The Scheduler's Wakeup Interval can be extended to reduce the frequency at which the Scheduler checks the Database for fresh Jobs. This offers benefits if the Scheduler communicates with the Database over a slow or congested network connection. PC-Duo Enterprise ships with a Scheduler Wakeup Interval of 60 seconds.

Each Scheduler is responsible for polling the Database and extracting new Jobs defined with its Host name. The polling frequency is controlled by the Wakeup Interval Host property. The Scheduler puts the Job into a queue pending the Job start time. When the Job is started, the Job unpacks its Client List. The List may contain any mixture of Clients, Fixed Groups and Dynamic Groups. Note that Dynamic Groups are evaluated when the Job commences.

Database Path

If a Scheduler's Database Path is to be changed, Schedulers using that Database must be stopped, reconfigured with the new location and restarted with the new path.

Maximum Active Jobs

The Maximum number of running Jobs controls the loading on a Host, and prevents a system becoming overloaded, for example if it is brought back online and several Jobs are due to run. The default value is 2. Note that a value of 0 means an unlimited number of running Jobs.

To configure WinScheduler:

1. Display the WinScheduler Properties dialog.
2. If the Scheduler is running, click the Stop button.
3. Modify the settings.
4. Click the Start button. This restarts the WinScheduler with the new settings.

To configure the Scheduler Service using WinScheduler:

1. Start the WinScheduler, then double-click on its icon in the taskbar's system tray to display the WinScheduler Properties dialog.
2. Click the Stop button if the WinScheduler is running. The settings can now be edited.
3. Modify the settings as appropriate.
4. Click the Start button to check and save the new settings.
5. Click the Stop button. Use the Run Service button to restart the Scheduler Service with the new settings.

   If you want the Scheduler to access a database stored on a file server at the same time as the user accesses the file server, the Scheduler Service must use the same logon account as the user. Otherwise, Scheduler activity could interfere with the user, and vice versa.
   
   The account used must have the appropriate networking privilege (such as Administrator) for the Scheduler to work. The Scheduler's account and password can be modified by selecting the Site, Scheduler Service Account menu command in the Console. The Scheduler must restarted to run with the new account information.

Submitting Jobs

You can submit an operation as a scheduled Job, and schedule it for a specific date and time using the Submit Job dialog.

To Submit a Job:

1. In the PC-Duo Enterprise Console, right-click on the operation you want to submit as a Scheduled Job. Choose Submit Job from the shortcut menu.

   The Submit Job dialog is displayed. It contains the following tabs:
   
   

   Tab	Description
   Job	Lets you name the Job and view the operation details.
   Schedule	Enables you to specify when and how often the Job is to be performed.
   Clients	Lets you drag and drop Nodes and Groups into a target list on which the Job will be performed.
   Logging	Enables you to control how a Job logs its progress.
   Retry	If the operation supports retries, use this tab to specify if and how the Job should be reattempted.

   
   

2. In the Submit Job dialog's Job tab, enter an appropriate Job name.
3. In the Submit Job dialog's Schedule tab, specify the Starting run date and Starting run time, then select the appropriate Host.
4. The Host drop-down list displays only those hosts which can support the submitted operation. Note that each operation can have a default Host defined, so that an appropriate Host is pre-selected for those operations when they are submitted. You can override this default selection if necessary.
5. Use the Type section of this tab to set the frequency of this scheduled Job. You can specify times that a Job is to be Scheduled, and a reschedule interval if required. Drag the mouse over the hourly display to set the reschedule times. If a repeating Job completes outside of the specified period, it is rescheduled for the first hourly slot of the next day.

   The Clients tab in the Submit Job dialog displays any Nodes or Group that were selected when you submitted the Job. This can include any combination of Clients and Groups. The Client List is unpacked by the Host each time it runs the Job. In this way, the PCs belonging to any Dynamic Groups are correct each time the Job runs.

6. In the Logging tab, select the appropriate options to specify how the Job should record its progress.
7. If the submitted operation supports retries, use the Submit Job dialog's Retry tab to specify how often the Host can attempt to start this job. The retry interval can be specified in units of Minutes, Hours, Days, or Weeks. This facility caters for online Clients that are turned off.
8. Click the OK button to submit the Job.

   A Job definition refers to an operation by name. The operation details are reloaded from the Database each time a Job runs so that updates to an operation are implemented by all the Jobs based on it. Any changes to the operation are picked up by any submitted Jobs that use the operation.

Specifying Tokenized Job Log Names

When specifying the Logging information in the Submit Job (and Modify Job) dialog, you can include tokens in the Log file name.

The default job log file name is $OPERTYPE_$HOST_$DDMMYY.LOG.

If you want to specify an alternative log file name, you can use the following tokens:

• $OPERTYPE - The short operation type name, such as 'HWINV' for Hardware Inventory.
• $JOBNAME - The name of the Job as specified in the Submit Job dialog's Job tab.
• $RETRY - The Job retry number. (A value of 1 is used for the first retry.)
• $HOST - The name of the Scheduler performing the Job.
• $MAXRETRY - The maximum number of retries. You could include this with the $RETRY token to create Job Log files such as HWINV_1_OF_5.LOG, HWINV_2_of_5.LOG, HWINV_3_of_5.LOG.
• $USERNAME - The Username used by the Scheduler.
• $FROMHOST - The name of the Console which submitted the Job.

Length-Restricted Tokens

There are also length limited versions of some of the tokens. These pass the first n characters of the token, where n is a number. For example, if you use a token of $JOBNAME(6) and the full Job name was MySoftwareInventory, only MySoft is passed:

• $OPERTYPE(n)
• $JOBNAME(n)
• $HOST(n)
• $USERNAME(n)
• $FROMHOST(n)

Date and Time Tokens

Various date and time formats can also be used as tokens in a Job log file name. These date and time tokens are evaluated when the job is run, so they reflect the actual job run date and time:

• $DATE
• $TIME
• $HHMM-Hours and minutes, not separated by a colon character and therefore suitable for use in the Job Log file name.
• $DDMMYY
• $MMDDYY
• $YYMMDD

Important Notes

If the tokenized file name string does not contain the $RETRY token, the retry number is inserted into the log file name. For example, a specified Log file name of HWINV.LOG becomes HWINV_1.LOG for the first retry.

The $TIME token can cause problems in a file name due to the colon (:) character. If you want to include the time in the Job log file name, we recommend you use the $HHMM token instead, which omits the colon character.

Monitoring Jobs in Progress

You can monitor the progress of Jobs using the Jobs Database window. This window consists of two sections, the Job Details (in the top part) and the Hosts View (in the bottom part).

The Jobs Database is the starting point for all Job management. The column headings in the Jobs window are mostly self-explanatory, but note that the Type field refers to whether the Job is to run once-only or repetitively.

The bottom part of the Jobs Database window is known as the Hosts View, and lists the Schedulers which are known to this Site Database.

To display the Jobs Database, click on the Show Jobs button in the Console toolbar. Alternatively, right-click in the Operations View of the Console, then choose Show Jobs/Hosts from the shortcut menu.

Working with the Hosts View

To display a Host's properties, right-click on the Host and choose Properties from the shortcut menu.

The Host Details dialog is displayed, showing the Host name, status, start time and number of Jobs processed. You can change some of the Properties for the Host in this dialog.

Click OK to return to the Jobs Database window.

Viewing Job Logs

Jobs post Client-by-Client progress information back into the Database. This progress information can therefore be viewed on any Console at the Site, in the Jobs Database window.

As well as static information, such as start time, each Job entry also shows the name of the PC currently being processed and the number of PCs processed so far.

The Job also creates a Log File in the working directory on the Host system, containing detail corresponding to the options specified in the Logging tab of the Submit Job dialog.

Job Log Files

Essentially, the Job Log file is a troubleshooting log file, which is not normally required. The default name of this file is PC-Duo Enterprise.LOG. Only the file for the latest Job is available. The first set of options in the Logging tab offers levels of information to be included in the file.

The Job Log file is not the same as the Scheduler's own system level log file, which contains top-level information on the status of the Scheduler itself.

The Logging tab of both the Submit Job dialog and Modify Job dialog features two options which are relevant to all Schedulers: Dummy run and Generate status report:

Dummy run tells the Scheduler to do everything in the Job except the per-PC processing. A dummy run unpacks the Client List, opens the Database, cycles through the Client List, and completes generating a Job Log, a Client Log and normally an empty Event Log. This can be useful if you want to check the actual Host commands that will be executed by the Job.

Generate status report provides a summary Client list at the end of the Job Log File. In effect, it reproduces the content of the Client Log in the Job Log text file.

Client Log

You can access the Client-by-Client success/fail status information using the Client Log. To open the log, right-click on a Job entry in the Job Log Database window, and choose View Client Log from the shortcut menu. The Client Log is displayed in the Client Report Log dialog.

The information in the Client Log enables you to review the overall success of the Job and enables you to confirm which PCs were processed and which had problems. The Client Log can be printed using the Log to File button, which also enables you to append the log to an existing file.

Event Log

The Event Log holds important Change and Exception information detected during the Job. For example, the detail of free space warnings, and summaries of changes detected during a Hardware or Software Inventory Job.

To display the Event Log, right-click on a Job entry in the Job Log Database window, then choose View Event Log from the shortcut menu.

Job Abort events are also shown in the Event Log, but other failures, such as failing to access a PC that is turned off, are held only in the Client Log.

You can use the Event Viewer's shortcut menu to save the Event list to a file. This is recommended if you need to save an Event report where truncation is a problem.

Deleting a Job Log also deletes the Client and Event Logs for that Job.

Scheduler Log File

The Scheduler maintains its own log file of Scheduler-level events. This log file typically records any fundamental errors which are detected, such as losing and recovering access to the Database.

This file is called LU32SCHEDULER.LOG by default, and is located in the PC-Duo Enterprise installation directory. It is only ever required for troubleshooting.

Modifying Jobs

PC-Duo Enterprise enables you to modify Jobs that are not presently running.

To modify a Job:

1. If the Jobs Database window is not displayed, click on the Show Jobs button in the toolbar.
2. In the top part of the Jobs Database window, right-click on the Job you want to modify, then choose Modify from the shortcut menu. The Modify Job dialog is displayed.

   You can now use this dialog to amend many aspects of the Job, including the Job name, the Clients on which the Job will be performed, and the scheduling information.

3. If you have defined a Job, but are not sure when to run it, defer it by setting a suitable first run date, such as Jan 1st 2010.

   The Job stays in the Database as Pending, and can be recalled for rescheduling when required.

Aborting Jobs

Use the Abort command in the Jobs Database window to abort a running Job. The Job is terminated when it finishes processing the current PC, and comments are posted in both the Job Log file and the Client Log section of the main Job Log.

Holding and Releasing Jobs

The Job Hold and Release commands in the shortcut menu allow you to delay execution of a pending Job indefinitely, and to then release it so that it executes when its scheduled start time becomes due.

Jobs that generate unusual errors during execution can also be placed into Holding state automatically. This enables you to review log information and to investigate and correct problems, before releasing the Job to run again.

Rescheduling Jobs on Completion

The Scheduler reschedules each Job when it completes. The Job is either put back in the Scheduler's queue as Pending its next run, or deleted if it was defined as Once-only.

Each time the Job is re-run, it is brought in from the Database and unpacked again. The entire Job is re-initialized, so any modifications made since the Job's last run are implemented.

Any changes to the Operation, or the Client List, Logging Levels, come into effect the next time the Job runs.

Deleting Jobs

The shortcut menu's Delete command enables you to delete the selected Job definition. Any associated log information generated by previous Job runs is also deleted. This option is available only for Jobs that are in Pending or Holding states.

Creating Database Operations

In addition to the Database Viewer, PC-Duo Enterprise also includes a utility that enables you to maintain and update your Site database. Use it to create customized operations that make your own SQL routines available from the Console.

WARNING: Take care when using this tool. DButil can overwrite data, delete tables and compromise the integrity of your database. Always test new DButil Operations on a copy of your database before using them with live data.

To create a DButil Operation:

1. Right-click on the Operations view and choose New in the Shortcut menu.
2. In the Add New Operation dialog, select Command Launch in the Type list and click OK.
3. In the Customize Command Launch dialog, enter an Operation name, and select the Operations view tabs in which you want to display the operation.
4. Select the Command Launch tab and type DBUTIL.EXE in the Program field.
5. In the Command line field, type:

$DSNACCESS "PROMPT=UserMessage" /SQLUPDATE "SQL=SQLStatement"

where:

$DSNACCESS is a token that specifies the DSN connection to the Site database.

"PROMPT=UserMessage" specifies the text of the user prompt that is displayed before the operation is run. Omit this parameter, if you want this operation to run with no user intervention.

/SQLUPDATE "SQL=SQLStatement" specifies a single SQL command, or the name of a command file. The file can contain multiple SQL commands and must be located in the Console directory.

DButil supports the Enterprise tokens in addition to SQL commands. For more information, see "Using Tokens in Command Launch Operations".

6. Choose the Run once only option and click OK to create the operation. You can now run the operation from the Console or submit it as a job.

   For examples of how to use DButil in the PC-Duo Enterprise Console, see the purge operations in the Utilities tab of the Operations view.

Creating SQL Command Files

Using a separate file allows DButil to execute several SQL commands in one operation. You can create a command file using a text editor, such as Notepad. Each SQL command must be entered on a separate line and must end with a semi-colon.

For example, to add an operation that changes the name of a Client PC in the database, create a file called ChangeNodeName.sql to update all Site database tables that contains the NODENAME field. For each database table that contains the name, use the following format:

 UPDATE Table SET NODENAME='NewNodeName' WHERE NODENAME='OldNodeName';

Save the file in the Console directory and create an operation that references ChangeNodeName.sql in the Command line.

Vector Networks http://www.vector-networks.com Voice: +44 (0) 1827 67333 Fax: +44 (0) 1827 67068 info@vector-networks.co.uk