Forecasting options and the new *INTERNAL forecast

A Robot/SCHEDULE forecast shows you the run activity scheduled for all Robot/SCHEDULE jobs over a time period you specify.

To make its forecast, Robot/SCHEDULE uses the job schedule, job completion history, and any information you add for user jobs, remote prerequisites, and members of remote groups. A forecast also can include a run-time analysis of each active subsystem. The run-time analysis uses the subsystem’s description to predict job queue wait time.

Creating a forecast
To build a forecast, simply specify the time period for the forecast. In Robot/SCHEDULE 10.0, you can specify the time period two different ways:

• Forecast Length lets you choose a number of days or hours. (This was the only option available in previous versions of Robot/SCHEDULE.) This forecast starts the moment you submit it and goes forward the specified number of days or hours. Jobs that are currently running and their reactive jobs are included.
• Forecast Period lets you enter from and to dates and times. With this option, the jobs included in the forecast depend on the date and time values you select. Group control and prerequisite jobs must be within the forecast period or their members/reactive jobs will not be included in the Forecast.

For example, suppose you have a job that runs at 8:00 on Mondays, with an average run time of 60 minutes, and another job that reacts to the successful completion of the 8:00 job, usually starting about 9:00. If you create a forecast using a Forecast Period that begins at 8:30 on Monday, the reactive job will not be included in the forecast because the 8:00 job (its prerequisite) is scheduled outside of the Forecast Period.
With either option, you can schedule the forecast build in Robot/SCHEDULE or submit it to batch immediately.

OPAL and schedule forecasts
Robot/SCHEDULE jobs that use OPAL might perform differently when they run compared to when the forecast was created.

For example, suppose you have OPAL code that checks the status of a job queue and skips the job if the job queue is held. If the job queue is held at the time the forecast is built, the job is not included in the forecast. However, when the job actually runs, the OPAL is evaluated. If the job queue is not held, the job will run.

Subsystems and schedule forecasts
At the bottom of the Create Forecast panel is a list of active subsystems. Forecasts automatically include all jobs from all subsystems, even those that do not appear on this list.

Robot/SCHEDULE uses this list to identify the subsystems for which it should perform run-time analysis. Run-time analysis attempts to determine job queue wait time to arrive at a more accurate start time for a job.

For example, suppose a job is scheduled to be submitted to a single-threaded job queue (only one job can be active at a time) at 4:00, but the forecast shows it starting at 4:45. Through run-time analysis, Robot/SCHEDULE discovers that another job that normally runs for 55 minutes is scheduled to be submitted to that same job queue at 3:50. The forecasting process determines the job will be delayed by 45 minutes. However, if you remove the job’s subsystem from the list on the Create Forecast panel, the job is forecasted to start at 4:00. (Removing a job’s subsystem from the list doesn’t exclude the job from the forecast, it only prevents the build process from doing run-time analysis.)

Once the forecast is built, you can view it on the Schedule Forecast List panel. The schedule forecast list shows each job scheduled to run during the forecast period, its start time, and its expected run time. You can adjust the forecast by changing start and run times. This lets you perform “what if” scenarios without affecting your schedule. You can even update the actual Robot/SCHEDULE job record.

*INTERNAL forecast
Robot/SCHEDULE 10.0 automatically creates a special internal forecast called *INTERNAL every time it starts and at 12-hour intervals at 11:00 and 23:00. The *INTERNAL forecast covers the next 24-hour period and is designed to be used by the Schedule Activity Monitor (SAM) and Robot/SCHEDULE job monitors.

SAM provides a visual reference to see that your jobs are running as expected. Whether you’re new to Robot/SCHEDULE or are an experienced user, SAM can help you monitor your job schedule. You can use SAM to monitor all the jobs you have set up using Robot/SCHEDULE. SAM can display information about jobs that are running, waiting to run, completed, failed, or forecasted.

Job monitors let you monitor your job schedule for specific events, such as jobs that run too long, complete too quickly, or start later than their scheduled start time. You can specify the criteria for each job at setup time. You also specify what Robot/SCHEDULE should do if it identifies a job monitor event.

The Late Start job monitor uses the *INTERNAL forecast to identify jobs that start later than their scheduled run time. If you add or change a job for which you want to specify a Late Start job monitor, you must include the job in a new build of the *INTERNAL forecast before it can be monitored.

Because the *INTERNAL forecast is used by Robot/SCHEDULE, you cannot modify or delete it. However, you can build a new *INTERNAL forecast at any time. You can even print the *INTERNAL forecast from the Schedule Forecast print dialog.

Contributed by Jean Monson, Maintenance Software Engineer

People often want to know how current the data on their SEQUEL dashboard is. To answer this question, you can add a Run Date/Time or System Date/Time time stamp to the dashboard. This article offers a couple of different ways to do this.

Use a SEQUEL View for a System i time stamp
You can use a simple SEQUEL View to display the reserved values Current Date and Current Time from the System i:

SELECT current date COLHDG(”Run Date”), current time COLHDG(”Run Time”)

FROM sqlexec

In this View the date and time values appear on two lines. To merge these into a single time stamp field, use the CHAR (character conversion) and CAT (concatenation) functions.

SELECT cat(char(current date) , ” - - “, char(current time))

NAME (time stamp) COLHDG(”Run Date/Time”)

FROM sqlexec

This displays the time stamp on one line as a single field.

Save the View, then add it to the dashboard using the Insert > SEQUEL Object menu option. After you’ve added it to the dashboard, you can drag it to any location and use the masking technique (explained later) to hide the heading and borders.

Use JavaScript for a desktop time stamp
Another approach is to use a text editor such as Notepad to create a simple Web page (.html file) containing the following HTML and JavaScript code:

<html>

<body>

<script type=”text/javascript”>

document.write(Date())

</script>

</body>

</html>

This code produces a time stamp in a browser window (shown below).

To improve the appearance, you can add HTML <font> tags and formatting specs around the <script> tags to change the font color, style, and weight, as shown in the following example.

<html>

<body>

<font color=”#000099″ size=”3″ face=”Arial, Helvetica, sans-serif”><b>

<script type=”text/javascript”>

document.write(Date())

</script>

</b></font>

</body>

</html>

Save the HTML file in a location on your Internet or network server.

Adding the time stamp to your dashboard
Open a dashboard in SEQUEL ViewPoint, or create a new one by choosing Select File > New > SEQUEL Dashboard. (The dashboard must be in design mode to add or modify objects. You’ll know you’re in design mode when the Insert menu option appears in the menu bar.)

Click on the Insert > Web Page option to add the new time stamp page. When the Web browser window displays, enter the Internet or network server address of the file you created earlier and click OK.

If you want the time stamp to show the last time the dashboard data was initialized or refreshed, use the default refresh settings. If you want the time stamp to function as a clock displaying the latest time, click File > Properties to display the Dashboard Properties window (below).

Select the Web page you just added. Then, click the Refresh Rate button at the bottom. Enter a 1 to refresh every second and click OK.

Formatting tip
If you want to display the time stamp window (or other dashboard objects) without the title and scroll bars, you can mask them. Click the Insert > Text option to display the Dashboard Text window. Enter some text in the Text box, then set the Text Color and Background Color to the same selection. You probably want this to be the background color, but initially it may be useful to pick a contrasting color. Uncheck the Show Border check box.

Create as many “blank” text boxes as you need. Next, resize and arrange them around the object you want to mask. Finally, edit the text boxes to match the background color.

Contributed by Jill Martin, Technical Services Manager

Now provides network-wide job schedule control

Help/Systems, developer of the Robot Automated Operations Solution for the System i, announces an enhancement to Robot/SCHEDULE 10, the automatic computer operator, job scheduling, and batch management software. Now, Robot/SCHEDULE interfaces directly with the Robot/NETWORK 10 graphical user interface to allow users to manage the job schedules on any system in their network; receive statuses from other systems; and provide cross-system reactivity.

Users define their network Host systems and their Nodes during Robot/NETWORK setup. When a Robot/SCHEDULE user connects to the Host system, the Robot/SCHEDULE Explorer tree displays the Host and its Nodes. The Robot/SCHEDULE Product Master on the Host system allows users to define jobs once and then distribute them and their associated objects to the Node systems or other Hosts. A new Send To Wizard walks the user through the send process, making it easy to distribute jobs, scheduling objects, and setup instructions to multiple systems. The ability to manage jobs from the Product Master helps users standardize and control their network job schedule.

Users can access the Node systems from the Explorer, allowing them to work on the systems as if they had connected to them directly. Menus and toolbars provide fast access to network control options—for the Host system, the Product Master, or the Nodes.

Robot/SCHEDULE’s consolidated reporting option allows users to consolidate selected data from multiple Nodes into a single report. Users can use the consolidated reports for comparisons of job processing, cross-system reactivity, and security events from multiple systems in the network.

You can view the Status Center many different ways

Robot/NETWORK 10.0 represents a big advance in network management. Let’s take a look at some of the major changes.

Robot/NETWORK Explorer
The Robot/NETWORK Explorer, a new PC interface, replaces the previous GUI and green screens. In Robot/NETWORK 10, you access all product functions through the Explorer (except for a few Product Master features).

The Explorer uses an expandable tree format to display all the Nodes for each Host. Click a Node to work with the Robot products installed on that Node. Right-click a product to view the Status Center for that product on that Node.

Right-click a Node to view the Status Center for that Node, stop or start the Node, or view jobs and output queues and system status for each Node. And, if needed, you can Telnet directly to the Node from the Explorer.

You can view the Status Center a number of different ways, depending on where you select the View Statuses option in the Robot/NETWORK Explorer. You can display the statuses for Hosts, Nodes, Robot products, or applications. Each status (for that product and Node) has a severity level that you can use to sort the statuses. You also can use a new filter option to sort the Status Center display.

Now, you can access Status History through status properties. These show when a status arrived on the Node, when it arrived on the Host, and when it was acknowledged, and by whom. If the event is a Robot/CONSOLE inquiry message, Robot/NETWORK also shows when, how, and who replied to it.

In addition to working with the Nodes and their products, you can use the Explorer to access Product Masters, packets, system setup, and other areas.

Map Center
In the Map Center, you see a representation of each system in your network and its current status. If you have multiple Hosts, you can view them all, or only a specific Host and its Nodes. You can tell at a glance if a system stops communicating with the Host, and you can stop or start it directly from the Map Center. By moving your cursor over a system, you see how many events are waiting for an action on a Node. You can access the Status Center for that Node directly to handle the events.

The Map Center also lets you view jobs, output queues, and system status for each Node and to Telnet directly to the Node to resolve issues.

You can choose the background display from a range of maps that come with Robot/NETWORK, or you can customize the background image using your own file.

Robot Product Maintenance
When you have several systems in your network, and a number of Robot products on each one, doing product updates can be quite a job. With the new Product Maintenance Wizard, you can download, distribute, and install Robot product updates for all your systems.

Robot Product License Manager
Use the Product License Manager to maintain your Robot product license codes across all your systems. You can key them in manually from the Product License Manager or request an XML file that contains license codes for all products on all systems. This XML file updates the codes across systems automatically. You don’t even need to sign on!

Consolidated Reports
Two new consolidated reports in Robot/NETWORK let you view information about events across the network.

The consolidated Good Morning Report displays status and packet information for all active Hosts and Nodes, over a specified date range. The report lists statuses by type and severity, by system. It also lists packets by status, user, and Node distributions.

The consolidated Status History Report displays the status history of events sent by Nodes to the Host system. You can select all Nodes and products, or a specific Node and product, for a date range.

Contributed by Anna Deuel, Product Manager

If I need to re-use a tape before its original expiration date, can I expire it manually in Robot/SAVE?
Yes, Robot/SAVE lets you expire a tape manually. Go to the Save Media Management Menu and select option 2 (By Volume Name). Locate the volume that you want to expire and enter a 9 next to the volume. A warning displays that the volume and its history will be removed. Press Enter to remove the history and expire the tape.

When you do this, you remove all the history for the objects saved on this tape. At this point, the data is still on the tape, but Robot/SAVE allows you to re-use the volume. The tape is reserved for the system on which you manually expired it.

I am trying to join two files using a derived field. SEQUEL ViewPoint displays the following error: “File xxx is not referenced in the JOIN specification.” How do I avoid the error so I can run the view?
Click OK to close the error message. Then, on the menu, select Options > ViewPoint Options > Design Tab. Un-check both “Ensure files are joined” and “Enable auto joining” in the File Join Options section. This allows the view to run with the derived join.

I have a Type 1 backup, with libraries that have objects to encrypt and libraries that do not. What is the most efficient way to set this up in Robot/SAVE?
To make your backup efficient, use one sequence number for all the libraries that do not have encrypted objects. Then, use another sequence number for the libraries that need encryption.

Is there a way for me to display just my monthly jobs in Robot/SCHEDULE?
It depends on how the jobs were set up. If your monthly jobs have a specific value in the application field, you can list them easily by doing a Robot query on this field. If your monthly jobs are scheduled using the DAYNO scheduling option, you could query that value. We recommend that when you set up Robot/SCHEDULE jobs you develop a naming convention that works for your installation. If monthly jobs are important, consider starting all job names with a dedicated character to identify them easily.