Configuring global settings for service health

As an administrator, you can define the health score computation mechanism, configure balancing profiles, and set both the health score and health status for all services from the Manage Service Health page.

Setting the service health score computation mechanism

BMC Helix AIOps offers the following mechanisms to compute the health score of a service:

• Node kind (default): Considers the node kind (database, host, virtual machine, and others), node weightage, and indexing to compute the health score of a service.
  For more information, see Health score computation by node kind.
• Impact severity: Considers the severity of the impact to determine the health score of a service. The score of the node with the highest impact (lowest score) is considered the service's health score. For more information, see Health score computation by impact severity.

To set the service health score computation mechanism for all services

1. Select Configurations > Manage Service Health.
2. Select from one of the following options:
   • Node Kind (default)
   • Impact Severity

Configuring balancing profiles

Global balancing profiles ensure that uniform resource thresholds are applied across all services. Applying the balancing profile across services minimizes noise by ensuring that a service's health is impacted only when the configured threshold is exceeded.

When there are a small number of services, adding a balancing profile to each service individually is manageable. As the number of services grows, this approach becomes inefficient and time-consuming. In such cases, you can create global balancing profiles from the Manage Service Health page. Based on the criteria defined during configuration, these profiles are applied to all relevant services. Global balancing profiles make sure that resource thresholds are consistently applied across all relevant services, helping you minimize disruptions, regardless of the number of services. To add a balancing profile for an individual service and to understand more about balancing profiles, see Adding-balancing-profiles.

Balancing profiles configured from the Manage Service Health page are applied to all services, including the child services that meet the defined criteria. This configuration includes the services that are created by using BMC Helix Discovery or any third-party application. The profile is applied only when matching CI Kinds exist within the service. You can define up to ten global balancing profiles from the Manage Service Health page. Balancing profiles defined at the individual service level takes precedence over the global profile.

To add a balancing profile for multiple services

1. Select Configurations > Manage Service Health.
2. Click Add Balancing Profile.
   The Define Balancing Profile pane appears.
3. Enter a unique profile name. 
4. Select a base CI Category and based on that, select the base CI Kind.
   The categories are displayed based on the CIs that are a part of the service.
   For example, you can define a global balancing profile for a Cluster, specifying that at least 80% of the hosts must be in an OK state, for the service to be healthy. This profile is applied to all the services of the Host CI kind that have a Host Containment relationship with the Cluster CI kind. These services are only impacted if the percentage of hosts in the OK state falls below 80%.
   
5. Select the following options:
   1. Link Kind: Select the type of relationship between the Base CI and the related CI nodes. 
   2. Evaluation scope CI Kind: Select the related CI kind.
      Based on the selection, the balancing profile is applied to the relevant CI kinds.
      For example, if Hosted Software Link Kind and Virtual Machine Evaluation scope CI Kind is selected, the balancing profile is applied to all CIs of a Virtual Machine that have a Hosted Software relationship with a CI Cluster. 
6. In the Fault tolerance policy section, select the quantifier by which you want to define the balancing rule:
   • CI Percentage: Specify the percentage threshold after which the service is impacted. 
     For example, if you specify 80%, the service is impacted when more than 20% of the resources stop working. A minimum of 80% of the CIs from the evaluation scope must be in the OK state. 

   • CI Count: Specify the number of resources after which the service is impacted. 
     For example, if you specify 20, the service is impacted if more than 20 resources stop working. A minimum of 20 CIs from the evaluation scope must be in the OK state. 

     Warning

     Important

     The health score of a service is displayed as 95 even if the threshold specified in the profile is not breached.

     For example, if the CI percentage is specified as 80%, the service is still displayed as healthy even if 10% of the resources are impacted, and the health score gets updated to 95. This informs the operators that the service while showing some impact, is not impacted enough to cause disruption.

7. Click Save.
   The balancing profile details are shown on the Manage Service Health page.

If you define the balancing profile from the Manage Service Health page, it appears in the Define Health Profile section for all services where the Use default configuration option is enabled.

In the example, the global profiles defined from the Manage Service Health page are visible in the Define Health Profile pane of this service. 

To edit a balancing profile defined for multiple services 

1. Select Configurations > Manage Service Health.
2. From the Service Health page, select the balancing profile that you want to edit.
3. From the Action menu for the profile, select Edit.

4. In the Define Balancing Profile pane, edit the profile as required.

   Warning

   Important

   If you add any CIs, the health score computation for the newly added CIs begins 15 minutes after the services are updated.

5. Click Save.

To delete a balancing profile defined for multiple services

1. Select Configurations > Manage Service Health.
2. From the Service Health page, select the balancing profile that you want to delete.
3. Select Action > Delete against the profile that you want to delete.
4. Click OK to delete the profile. 

Setting the service impact health score for all services

Use the Health Score Configuration section on the Manage Service Health page to customize the health score impact for all services. These global settings apply across all services, ensuring consistency and significantly reducing manual effort. If needed, you can override the global configuration for specific services by updating their individual health profiles.

To set the service health score

1. Select Configurations > Manage Service Health.

2. Expand the Health Score Configuration section. 

3. Enter the new numbers or use the up and down arrows to adjust them.
   For example, if you update the impact score value for a Critical event to 20, then for every critical event, 20 points are deducted from 100, and the health score is computed.
   You cannot specify fractions or decimal points for the score. 

4. Click Save and Close. 
   The health score for all the services is now computed based on the updated values. It may take a few seconds for the recomputed score to reflect across services. 

   

Setting the service health status for all services

Use the Health Status Configuration section on the Manage Service Health page to customize the health status for all services. The global settings you define here apply across all services, ensuring consistency and reducing manual effort. If needed, you can override the global configuration for specific services by updating their individual health profiles.

To set the service health status

1. Select Configurations > Manage Service Health.

2. Expand the Health Status Configuration section.

3. Enter the required numbers, or use the arrow keys to adjust the values.
   You cannot specify fractions or decimal points for the threshold. 

4. (Optional) To exclude severity from health score calculations, clear the Enabled check boxes.

5. Click Save and Close. 
   The health status for all services is now calculated based on the updated values. It may take a few seconds for the changes to be reflected across services.