2.2 Using the Orchestrate Server Portal

This section describes how to perform some basic tasks with the portal for PlateSpin Orchestrate. The following content is included:

2.2.1 Connecting to the Orchestrate Server Portal

Use the following steps to run the portal and log in at a workstation:

  1. Start any Internet browser.

  2. Enter the URL (server name) of the Orchestrate Server to display a page where the user can log in to the server:

  3. Specify the username and password created for you by the PlateSpin Orchestrate administrator.

  4. Click Log In to log into the portal.

The Orchestrate Server Portal Interface

After you log in to the portal, you see the following interface displayed in the browser:

Figure 2-1 The Server Portal Interface

NOTE:Logging in with the username of a PlateSpin Orchestrate administrator displays additional menu options for Server Status and User search.

The table below provides details about the various labeled areas of the interface.

Table 2-1 Server Portal Interface Areas

Area

Description

User Login and Group Roles

Indicates the user groups that the currently logged-in user belongs to and whether the user has an “administrator” role or a “user” role.

Main Menu

Provides operations categorized under the menus:

  • Jobs include a list of running and completed jobs on the Orchestrate Server and their status.

  • Run Job allows the user to submit a job to the Orchestrate Server, optionally specifying parameter values for the job.

  • Resource Search specifies which resources are to be selected for use by joblets associated with the jobs to be run.

  • Logout returns the user to the login screen.

Filters

Provides a way to filter the view of running and completed jobs.

View and Function

Displays jobs, computing resources, or other requested information. This is the general area of the portal for viewing requested data.

Commands

Provide a set of commands for the functions that can be performed on the data in the View and Function area.

Help

When expanded, displays context sensitive help information when the user mouses over some areas of the interface.

2.2.2 Displaying Available Jobs

After you log in, you can use the portal to display available jobs that can be run on the PlateSpin Orchestrate system.

To see runnable jobs from the portal command menus, click Run Job in the menu bar to open the Run Job page. The Run Job page allows you to specify what job to run, where the selected job’s joblets execute, and when the job runs.

The Select Job tab is listed in the View and Function area of the portal interface. The Available Jobs explorer is located on the left side of this interface. The Job Parameters area is located on the right side. The Job Parameters area is associated with job selections in the Available Jobs explorer. To view all jobs, expand the explorer under Available Jobs > all.

PlateSpin Orchestrate includes prepackaged jobs for demonstration. Prepackaged jobs are optionally deployed with each new server when it is first started. These jobs include demoInterator, failover, quickie, subjobs and so on.

Figure 2-2 The Available Jobs Explorer View Expanded to Show All Jobs

2.2.3 Running a Job

There are four stages to running an PlateSpin Orchestrate job on managed computing resources. This section uses the quickie example job to illustrate these stages:

NOTE:Quickie is a demonstration job used primarily to show computing resource utilization. Its purpose is to simply start its executable program on an appropriate computing resource in the grid, wait for a period of time, and then stop to free the computing resource again.

Selecting the Job You Want to Run

When you want to run a job, you need to first select the job and specify its parameters. The portal interface labels this as the What stage of running a job.

Use the following steps to select the job and specify job parameters:

  1. From the Select Job Explorer, expand the examples folder in the Available Jobs directory tree to list the example jobs that are deployed on the PlateSpin Orchestrate Server.

  2. Select the quickie job to display its job parameters in the Job Parameters area of the interface.

    If you need more information about a job parameter, click Desc to expand the description of the fields of the Job Parameters area.

  3. Accept the default values for the parameters.

    The numbjoblets parameter is an argument for the associated job. The default parameter value for the quickie job is 100.

  4. Click Lock Job to select the job.

    The What stage on the interface now lists the quickie job selection. If you want to change the job selection or its parameters, you can click Edit on this stage anytime before you actually run the job.

Selecting Computing Resources to Run the Job

After the job is selected, users need to identify the computing resources where the joblets associated with the job are to be run. The portal interface labels this as the Where stage.

Searching for computing resources is done through Resource Search. When the Resource Search page is displayed, users can select computing resource search criteria to search for the computing resources they want.

Your grid might have different kinds of computing resources, identified by a set of facts about each computing resource. Examples of resource facts include the resource name, hardware configuration, operating system, and other characteristics. If the grid has uniform resources, the Where stage is less useful, but for a grid with multiple resource characteristics, the Where stage is very important.

Depending on the resources that you have set up with the PlateSpin Orchestrate Server, you might have multiple computing resource operating system types (operating system fact). For example, you could have both Windows* and Linux operating system resources available. Some of the example jobs that come with the PlateSpin Orchestrate Server have executables for both operating systems, and others do not. The quickie job can run on both Windows and Linux.

The following sections contain more information about identifying computing resources:

Using Expressions to Identify Appropriate Computing Resources

A simple expression is used for identifying the computing resources appropriate for running a job. Multiple expressions can be applied. Each expression is entered by using the expression criteria listed in the following table.

Table 2-2 Expressions Used to Filter Computing Resources Eligible to Run Jobs

Expression

Criteria

Fact

Each computing resource has a set of facts. A complete list of facts is available from the Select Fact drop-down list for computing resources.

Operator

Each expression requires one operator.

Value

The value criteria are evaluated against the associated fact according to the selected operator. The portal generates a list of possible value options based on the selected fact and the possible values from the current set of resources.

When you select a value, it is entered into the criteria edit field to the right of the value selection field. You can edit the value in this field if necessary.

Multiple criteria can be entered in the same fashion. The selected computing resources match all criteria entered. You can alter your choice of expressions by using the following functions:

  • To remove a criterion, click the minus ( - ) icon located to the right of the criterion.

  • To start a new search or to start over, click Reset located below the criteria list.

  • To add more criteria, click the plus ( + ) icon.

When you have entered all of the search criteria, click Search. The matching resources are listed on the lower part of the page under Search Results.

Sample Job: Selecting Computing Resources for Running a Job

If no search facts are selected, PlateSpin Orchestrate assumes that all computing resources are to be selected.

The following steps demonstrate how the quickie job is configured to run only on Windows XP operating system computing resources:

  1. From the Select Fact drop-down list, scroll to and select the operating system fact resource.os.name.

  2. From the Select Operator drop-down list, specify the operator to use for the fact. In this case the default, is equal to, is used.

  3. From the Select Value drop-down list, specify the fact value, Windows XP.

    Each of the drop-down items is context sensitive. When a fact is selected, the values update with the valid options for that fact, based on the current set of available computing resources. The values change automatically as computing resources are added or removed in the grid of managed resources.

  4. Click Search to check whether any computing resources match the current search filter criteria. If so, these computing resources are listed at the bottom of the Function area of the portal. See the figure below for an example.

  5. Select the check box next to the desired resources, then click Lock Criteria (located at the bottom right of the Function area of the portal) to select the computing resources.

Notice that the Where section (located at the top of the Function area of the portal) now shows the quickie job resource selections. To change the computing resource selection, click Edit before running a job.

Scheduling a Time to Run the Job

After you have selected a job and the computing resources to be used, you need to select a start time for running the job. The portal labels this as the When stage of running a job.

The Reservation tab area shows start and end requirements, as well as a priority setting for running the job. In the When stage, the user can specify a start and end time and a job priority for the selected job.

Use the following steps to specify when the job is to be run.

  1. Click the Reservation tab, click the calendar icon located adjacent to the Start field to open a calendar scheduling utility, then select a date and time to run the job. You can also accept the default in the Start field: As Soon As Possible.

  2. (Optional) Click the calendar icon located adjacent to the End field, then select a date and time to complete the job. You can also accept the default in the End field: Job Duration.

  3. (Optional) In the Priority field, click the drop-down list to select a priority for the job, or accept the default.

    Use the Priority parameter to increase or decrease the priority of the job in relation to other jobs that are already running or scheduled to run. This option is used only if computing resources are inadequate to run the job at the scheduled job start time. The options in the drop-down list are context-sensitive and match the available priorities associated with the user profile.

  4. Click Lock Time (located at the bottom right of the Function area) to finalize the time to run the job.

The job is now ready to run. The job criteria are displayed in the upper part of the Function area of the portal.

Figure 2-3 The Server Portal Displaying the Job Run Criteria Previously Selected

Running the Job

When you have established the job run criteria in the Server Portal, click Go to schedule the job to run. A progress log is displayed in the Progress tab in the Function area of the portal.

Figure 2-4 Progress Message Displayed for the “Quickie” Job

A Done message indicates that the job was successfully submitted to the server for PlateSpin Orchestrate.

Click OK to close the Progress log and return to the Job Status page to monitor the running job.

NOTE:This same job could also be run using the following zos command:

zos run quickie.job

For more information about using the zos command line interface, see The zos Command Line Tool in the PlateSpin Orchestrate 2.0 Command Line Reference.

2.2.4 Monitoring a Job

Users can access a list of PlateSpin Orchestrate jobs that are currently running by clicking the Jobs command menu.

Figure 2-5 Jobs Command Menu on the Server Portal

The Job Status page lists information about jobs submitted to PlateSpin Orchestrate.

Figure 2-6 Job Status Page Showing the Job Filter Drop Down List

A list of jobs is displayed based on the filter criteria selected. The default filter uses all jobs for the currently logged-in user. Filter options include:

  • Users: Select a user, or one of the built-in group types: group_all for all user groups, or group_system for all system jobs.

  • Job: Select a specific job, or the built-in group types: group_all for all jobs.

  • History: Select jobs submitted within the last nn hours or nn days from the current date and time. The minimum time you can set is 1 hour. The maximum time you can set is 24 days. If you select none as the history time, only the most recent jobs are selected.

When you have selected the filter criteria you want, you can update the currently displayed list. Click Refresh (located at the bottom right of the interface).

Figure 2-7 Status Information for a Current Job

The job status report includes the following columns:

  • The left column indicates the result for the job status report as sampled at refresh time. The column displays the result with both an icon and text indicators. The job result types are Canceled, Scheduled, Running, Failed, or Completed.

  • Identification: This column lists job submission information, including User, Job, Name, and Priority.

  • Schedule: This column lists information about when the job is scheduled, or its actual start time. It also includes how long the job has been running. If the job is completed, the column also includes the total elapsed run time for job.

  • Status: This column lists information about the cost and progress of a job. For jobs that are currently running, the progress indicator bar is partially filled with the percentage of job completion. Cost indicates the current cost utilization for a running job, or the total cost for completed and failed jobs. The cost of a job is based on facts configured by administrator for various resources.

  • Misc: This column lists other useful information about a job, including its Subjobs, its Joblets (the number of independent pieces of work running on each computing resource), and any Errors.

    If the number of subjobs is greater than 0, a view… link displays in this column. Click this link to show a Job Tree hierarchy under the Filter settings for the job. All of the subjobs in the job hierarchy are shown.

    Figure 2-8 Job Tree Hierarchy Displayed on the Job Status Page

    Click Root in the tree to return to the default Job Status display, which is based on the current filter criteria.

Using the Command Line Interface to Monitor Jobs

Use the following command if you want to use the command line interface to monitor the job: 

zos status

By default, PlateSpin Orchestrate returns the status of the last job to run.

For more detailed information about using the zos command line interface, see The zos Command Line Tool in the PlateSpin Orchestrate 2.0 Command Line Reference.

2.2.5 Viewing the Job Log

From the Jobs menu in the portal interface, some jobs display a log… link on the Identification column of the Job Status page.

Figure 2-9 Job Identification with an Available Log Link

Selecting this link displays the logging information for the job. The logging display stays current for the job shown running. When the job is completed, its complete log is displayed.

2.2.6 Pausing and Resuming a Running Job

From the Jobs menu in the portal, display the job to pause or resume by including it in the Job Status list using the filters. Use the check box on the left side to select the running job to pause or resume. Then click Pause or Resume on the bottom of the page. A new page displays, indicating the status of the request. Click OK to return to the Job Status page.

2.2.7 Canceling a Running Job

  1. From the Jobs menu, display the job to cancel by including it in the Job Status list using the filters.

  2. Select the check box for the running job you want to cancel.

  3. Click Cancel.

    A new page displays, indicating the status of the cancellation request.

  4. Click OK to return to the Job Status page.

2.2.8 Logging Out of the Orchestrate Server Portal

To log out of the Orchestrate Server Portal, click the Logout option under the Main Menu.