How to Provide Administrative Data

Orientation
What Data is Needed?
What Data is Already Available?
Deployment Topology Considerations
Life Cycle for Administrative Data
1. Start Autoload
2. Feed Data to Autoload
Audit Data
Relationship Data
Agent-Collected Metrics

Orientation

What Data is Needed?

Before data is available in an administrative report, the data must be collected or generated by a source system, extracted to a drop zone, and autoloaded to memory. To determine which data source is used by a particular report section, open the report in the designer, select the report section that you are interested in, and then select the Data tab in the left pane.

What Data is Already Available?

  1. In the administrator, select LASRthen selectManage Tables from the main menu.
  2. Right-click on any column heading, and make sure the LASR Name column is selected.
  3. In the tab toolbar, select LASR name from the drop-down list, and enter EVDM in the search field.
  4. Check the check box in the first column heading so that all tables are selected. Then, click the get status icon in the tab toolbar. To interpret the display, see Get Table Information.

Deployment Topology Considerations

Support for administrative data is limited as follows:
  • If the middle tier and server tier are deployed on separate machines with dissimilar operating systems (one tier on Linux and the other tier on Windows), the audit and relationship data feeds are not supported.
  • If the middle tier and server tier are deployed on separate machines with similar operating systems (both on Linux or both on Windows), the autoload data directory (drop zone) for administrative data must be available to both machines.
    Windows Specifics: Share the directory. Change the EVDMLA library’s VA.AutoLoad.Location extended attribute to be a UNC path.
    UNIX Specifics: Put the directory in an NFS-mounted location. Change the EVDMLA library’s VA.AutoLoad.Location extended attribute to point to the NFS-mounted location.

Life Cycle for Administrative Data

For audit data, see Key Actions Auditing.
You can periodically retire other tables by deleting them from the autoload data directory (/AppData/SASVisualAnalytics/VisualAnalyticsAdministrator/AutoLoad/EVDMLA) or moving them to a backup location.

1. Start Autoload

To start autoload for administrative data:
  1. On the machine that hosts the administrative reporting library, identify or create a scheduler account.
    • Give the account the host-layer privileges that are required to start the associated SAS LASR Analytic Server and load data. See Host Account Privileges.
    • On UNIX, enable the account to run cron jobs.
    • In the SAS configuration directory, give the account Read and Write access to the following autoload directories and their contents:
      Data:
      /AppData/SASVisualAnalytics/VisualAnalyticsAdministrator/AutoLoad/EVDMLA
      Scripts:
      /Applications/SASVisualAnalytics/VisualAnalyticsAdministrator/EVDMLA
  2. In the metadata, create a corresponding individual metadata identity. See Adding Users.
    Note: This requirement reflects the standard configuration. See Metadata Server Connection.
    Give the scheduler account’s metadata identity the required metadata-layer permissions on the target server, library, and folder. A simple approach is to add the scheduler account’s metadata identity to the Visual Analytics Data Administrators group. An alternative is to grant access to the metadata identity as follows:
    Server:
    LASR Analytic Server
    RM, WM, A
    Library:
    Environment Manager Data Mart LASR
    RM, R, WM, A
    Folder:
    /Shared Data/SAS Visual Analytics/Autoload/EVDMLA
    RM, R, WMM, W
  3. Log on to the host as the scheduler account, navigate to the scripts directory for EVDMLA, and invoke schedule.sh (on UNIX) (or schedule.bat (on Windows).
    Tip
    The schedule script that is in the VisualAnalyticsAdministrator directory starts a different library’s implementation of autoload. You must invoke the script that is in /VisualAnalyticsAdministrator/EVDMLA. See Autoload.
  4. Verify that the scheduled task is running.
    Windows Specifics: Access the Task Scheduler (for example, select Startthen selectControl Panelthen selectAdministrative Toolsthen selectTask Scheduler). Locate the task Environment Manager LASR - Auto Load Scheduler in the Task Scheduler Library.
    UNIX Specifics: Run the command: crontab -l

2. Feed Data to Autoload

To efficiently enable multiple data feeds, defer server restart and verification steps until all configuration property changes have been made.

Audit Data

To feed audit data to the drop zone:
  1. Initiate generation and periodic extraction of audit data. See How to Safely Enable Auditing.
    Note: In addition to starting data collection, this step extracts certain audit records from the audit service’s database, and feeds that data to the Append directory in the administrative reporting drop zone.
    Note: Until autoload is started, the data feed occurs only one time.
  2. Perform some tasks that generate audit records. For example, load or import a table, or create and save a report or exploration.
  3. After 30 minutes, verify that the LASR table EVDM.AUDIT_VISUALANALYTICS is loaded.
Here are some details:
  • Audit data is provided by the audit service. See Audit Content and Coverage.
  • Extraction occurs only if the data builder is licensed, installed, and running.
  • Extraction uses the pooled workspace server in the suite-level default SAS Application Server. See va.defaultWorkspaceServer.
  • The extraction process must have Read and Write access to the autoload data directory and that directory’s Append subdirectory.
    Note: Autoload’s append action is used, so the extracted table is written to the Append subdirectory.
  • If autoload is not running, the data feed to the drop zone occurs only one time.
  • A separate scheduled task manages the size of the full AUDIT_VISUALANALYTICS table in the drop zone. See About auditRefresh.

Relationship Data

To feed relationship data to the drop zone:
  1. Enable collection and extraction of relationship data.
    1. To set up scheduled data collection (or verify that data collection is already scheduled), see Configuring Automatic Relationship Loading and Cleaning in the SAS Intelligence Platform: System Administration Guide.
    2. To enable periodic extraction, set va.extractRelationshipData to true. The extraction interval is 3 hours.
    3. Restart the SAS Web Application Server.
  2. After a sufficient interval has elapsed, verify that the LASR table EVDM.RELATIONSHIPS_VISUALANALYTICS is loaded.
Here are some details:
  • Relationship data is provided by the relationship service. See Using the Batch Relationship Reporting Tools in the SAS Intelligence Platform: System Administration Guide.
  • Relationship data is extracted for the following object types:
    • 2G reports (Report.BI objects)
    • Explorations (VisualExploration objects)
    • Data queries (VisualDataQuery objects)
    • Data sources (Table objects)
    Note: Relationship data is also extracted for objects that are directly referenced by any of the preceding object types.
  • The extraction process must have Read and Write access to the autoload data directory.
    Note: Autoload’s refresh action is used, so the extracted table is written to the autoload data directory.

Agent-Collected Metrics

Note: Agent-collected metrics do not provide information about distributed SAS LASR Analytic Servers.
To feed agent-collected metrics (ACM) to the drop zone:
  1. Enable data collection. See Initializing and Enabling the Service Management Architecture in the SAS Environment Manager: User’s Guide.
  2. Enable data transfer to the EVDMLA drop zone. See Feeding Data From the Data Mart into SAS Visual Analytics in the SAS Environment Manager: User’s Guide.
    Tip
    A supporting format catalog must be available. When you enable this data feed, the required format catalog is added to the /AppData/SASVisualAnalytics/VisualAnalyticsAdministrator/AutoLoad/EVDMLA/Formats directory. For some applications, the format catalog must also be added to the path of the appropriate SAS Application Server. See Supporting User-Defined Formats.
  3. After a sufficient interval has elapsed, verify that the expected ACM tables are loaded. For information about the source data, see ACM Tables in the SAS Environment Manager: User’s Guide. (The list of tables is subject to change in future releases of SAS Environment Manager.)
Copyright © SAS Institute Inc. All Rights Reserved.
Last updated: December 18, 2018