Thank you for your feedback.
Form temporarily unavailable. Please try again or contact to submit your comments.

Operational Intelligence release notes

Log in to subscribe to topics and get notified when content changes.

Operational Intelligence release notes

ServiceNow® Operational Intelligence product enhancements and updates in the Madrid release.

Operational Intelligence is an add-on application to Event Management, with a key preventive role in a service-centric solution for reducing service outages. With Operational Intelligence, you can proactively identify service issues, pinpoint service outages, and automate remediation.

Operational Intelligence captures raw metric data from various data sources such as the SolarWinds monitoring server and the Nagios XI server. Then, based on the analysis of the historical metric data, Operational Intelligence detects anomalies that Event Management may not capture. Operational Intelligence generates anomaly alerts that you can promote to regular Event Management alerts. Those alerts appear on the Alert Console and service health dashboard, and you can use Alert Intelligence in Agent Workspace to manage.

Madrid upgrade information

  • In Madrid, user-specified width override values are replaced by new advanced settings that are used internally to calculate width values. Width override values are preserved through an upgrade to Madrid and are being used internally. However, they do not appear in the UI and you cannot modify them. When you use the Bounds Settings wizard in an upgraded instance and you select a metric class that has a width override value, a notification appears. To use the new advanced settings in the Bounds Settings wizard, accept the notification to delete the width override values from an earlier release.

    For more information about the Bounds Settings wizard, see Custom bounds settings.

  • During the upgrade to Madrid, the Apache Ignite software that runs on Operational Intelligence MID Servers in MID Server distributed clusters is upgraded to version 2.5.3. After you upgrade all the MID Servers in the MID Server distributed cluster to Madrid, restart these Operational Intelligence MID Servers to complete the Apache Ignite upgrade.

    You might have some data loss during the time that the MID Servers upgrade starts until the MID Servers are restarted. To minimize this data loss, before you start the upgrade to Madrid, disable the metric connectors. After the upgrade completes, set the max_fetch_interval_min parameter to how long the connectors were down and then enable the connectors so that the missing data is pulled.

New in the Madrid release

Self-health monitoring
Monitor essential Operational Intelligence components and processes, and generate alerts about potential issues. Operational Intelligence uses pre-configured self-health monitors for Event Management which enables you to proactively remediate issues and minimize data loss.
Quick start tests for Operational Intelligence
Validate that Operational Intelligence still works after you make any configuration change such as applying an upgrade or developing an application. Copy and customize these quick start tests to pass when using your instance-specific data.

Changed in this release

When you initially set up Operational Intelligence, use the Setup Operational Intelligence related link on a MID Server form to automatically do the following:
  1. Configure a MID Server for Operational Intelligence
  2. Create a MID Server distributed cluster that contains the Operational Intelligence MID Server
  3. Create an Operational Intelligence Metrics Extension and start it
  4. Create the MID Web Server extension configured with the ‘Enable REST Listener’ option
Custom bounds settings

The Bounds Settings wizard enables you to specify custom upper and lower bounds for a metric. Custom control bounds override the learned control bounds for a metric, which affects the resulting anomaly scores. The Bounds Settings wizard also enables you to preview anomaly analysis of actual data using the new custom settings. This preview is similar to previews that are generated by anomaly model testing.

The Bounds Settings wizard automatically calculates statistical models, such as Static Centered or Static Skewed, and width values that are based on your settings. The Bounds Settings wizard also automatically creates the metric class and configuration setting rule. Therefore, there is no need to manually set or create any of those items.

Resource binding
Monitor resources, such as disks and processors, without creating CIs, which saves you the maintenance that those CIs would require. Resource binding also enables you to monitor an entity that is not a CI, such as a web page, which is an application.

Resource binding creates a resource record for each resource (such as a disk), which streamlines metric data and results in a meaningful aggregation.

Metric to CI mappings
View the status of metric to CI and resource mappings, including details about unsuccessful mappings. CI or resource mapping can fail because of a missing or duplicate CI, or because a resource table is missing. If Operational Intelligence cannot map a metric to a CI, then that CI is not included in anomaly detection until it is properly mapped.

Use the details provided for a failed mapping to investigate and remedy common mapping problems.

Anomaly detection
Exclude metrics for CIs that are in maintenance mode from model learning. When the system property sa.model_learner.maint_event_record_history_enabled is set to true (default), Operational Intelligence stores historical information about the times CIs enter and exit maintenance mode. When the system property sa_metric.maint_exclusion is set to true (default), metrics from CIs that are in maintenance mode are excluded from model learning. Change the settings of these properties to include or exclude metric data for CIs while they are in maintenance mode.
The Stationary Nonparametric statistical model creates control bounds that can fit actual data better than other statistical models. Once learned, the control bounds persist until the next learning cycle of incoming data.
User interface navigation
Operational Intelligence modules in the navigation bar are now logically grouped by function areas so that you can quickly find what you are looking for.

Activation information

Operational Intelligence plugin ( requires a separate subscription and must be activated by ServiceNow personnel. This plugin includes demo data and activates related plugins if they are not already active.