Configuring a monitoring source for IBM Operational Decision Manager events

For your Red Hat® OpenShift® deployment of IBM Business Automation Insights to enable creation dashboards for events from custom sources in Business Performance Center, you must configure the monitoring_sources subsection of the custom resource YAML file.

Before you begin

You must create a time series in OpenSearch before you can customize the processing application. Without the time series, the custom monitoring is not processed, and the cockpit pod logs show the following message:

The monitoring source 'MyCustomMonitoringSourceFieldExplosion' could not be loaded because the associated index 'icp4ba-bai-odm-timeseries1-ibm-bai' does not exist in the OpenSearch instance.

About this task

The custom resource (CR) file of the Business Automation Insights operator contains a business_performance_center section where you specify the settings for Business Performance Center. In the dedicated monitoring_sources subsection, you specify the definitions of the monitoring sources for Operational Decision Manager events.
Monitoring source definition
A monitoring source definition provides the information that is required to generate monitoring sources for events in Business Performance Center.
- monitoring_source_type: "<type>"
  opensearch_index: "<index name>"
  monitoring_source_fields:
    - field: "<field name 1 (mandatory)>"
      label_field: "<field label 1 (optional)>"
    - field: "<field name 2 (optional)>"
      label_field: "<field label 2 (optional)>"
Parameters
Set the placeholder values as follows.
monitoring_source_type
The type of the monitoring source. It can contain space characters.
opensearch_index
The name of the OpenSearch index.
monitoring_source_fields
In this section, you must specify at least one field, which is later used to retrieve and aggregate data in OpenSearch for your Operational Decision Manager events. You can also provide a field to serve as the label for the monitoring sources. If no label field is provided, the value of the field that is specified to retrieve and aggregate data is used as the label. If more than one field is required to identify monitoring sources for your Operational Decision Manager events, you can provide several pairs of field and label fields.
For example, consider the following monitoring source fields.
monitoring_source_fields:
  - field: "item_id.keyword"
    label_field: "item_name"
item_id
This field consists in a keyword type in OpenSearch which represents the identifier of an item. To specify the associated field with a keyword type in OpenSearch, you must append the .keyword suffix to the field name.
item_name
This field consists in a text type in OpenSearch that represents the name of an item.
One monitoring source exists for each distinct value of the item_id field. The corresponding value of the item_name field is used as a label for selecting the monitoring sources in Business Performance Center. If more than one versions exist of a given item with a given identifier, you can use an item_version field to specify one monitoring source for each version of item.
monitoring_source_fields:
  - field: "item_id.keyword"
    label_field: "item_name"
  - field: "item_version"
Note: In the monitoring_source_fields section, field elements must have the type keyword in OpenSearch whereas no type constraint exists on the fields that are used as label_fields.
For more information
About naming conventions for indices, see Naming conventions for indices and aliases.

Procedure

  1. Add or edit the business_performance_center section of the custom resource (CR) YAML file.
  2. Add or edit the monitoring_sources section under business_performance_center.
  3. In the monitoring_sources section, add a new monitoring source for your Operational Decision Manager events.
    bai_configuration:
      business_performance_center:
        monitoring_sources:
          - monitoring_source_type: "<type>"
            opensearch_index: "<index name>"
            monitoring_source_fields:
              - field: "<field name 1 (mandatory)>"
                label_field: "<field label 1 (optional)>"
              - field: "<field name 2 (optional)>"
                label_field: "<field label 2 (optional)>"
    
  4. Save the changes.
  5. Apply the changes.