Incident Management

Incident Management provides a centralized way to configure notifications and track incidents raised by policy executions across your organization. Instead of configuring alerts separately on each policy, you create Incident Policies that define which policies or rules to monitor, what condition should raise an incident, and how you want to be notified. Every incident raised this way is tracked in a single Incidents list.

To access Incident Management, navigate to Operations > Incident Management in the left navigation.

Incidents

The Incidents tab lists every incident raised by your Incident Policies.

Use the following controls to narrow the list:

The incidents table includes the following columns:

Incident Policies

An Incident Policy defines what to monitor, the condition that raises an incident, and how notifications are sent. To create one, navigate to Incident Management > Incident Policies and select Create Policy.

What do you want to monitor?

Select the entity type to monitor. Data Reliability is currently available; Pipelines, Jobs, and Data Products are shown as Coming soon.

Next, choose how to configure the policy:

• Configure by: Rules (monitor specific rules) or Policies (monitor specific policies).
• Global Condition or Individual Condition (see Setting conditions and severity below).

Selecting rules or policies

Select Add Rules or Add Policies to open a dialog listing the available rules or policies. Use the following controls to find what you want to monitor:

• Policy type quick filters: Data Quality, Reconciliation, Freshness, Anomaly, Data Drift, Schema Drift, and Auto Anomaly.
• Overall Status filter and More Filters for additional criteria.
• Search by policy name.
• Columns: Show or hide table columns, including Policy Type, Tags, Quality Score, Open Alerts, Overall Status, Rule Count, Records Processed, Started At, Ended At, and Total Execution.

Select the checkbox next to each rule or policy you want to monitor, then select Add Policies (or Add Rules) to confirm, or Cancel to discard your selection. Selected items appear in a list on the main page, each showing its policy type and name, with an option to remove it individually. You can return to this dialog at any time using Add Policies / Add Rules to add more.

Setting conditions and severity

How you configure the trigger condition depends on whether you selected Global Condition or Individual Condition.

Global Condition applies one shared condition and severity to every selected rule or policy:

• Condition: Choose Execution status or Result status, an operator, and one or more status values from Aborted, Errored, Successful, or Warning.
• Severity: Critical, High, Medium, or Low.

Individual Condition lets you configure a separate condition and severity for each selected rule or policy, shown as a table with Policy (or Rule), Condition, and Severity columns — one row per item, using the same Condition and Severity options described above. Select the X on a row to remove that rule or policy from the policy.

Notification timing

Choose how you want to be notified:

Depending on the option selected, configure:

• Where should we send instant notifications? — Select one or more notification channels. Shown for Instant and Instant + Summary.
• Where should we send summary notifications? — Select one or more notification channels. Shown for Scheduled Summary and Instant + Summary.
• When should we send summary notifications? — Set a Time zone and one or more Schedules (for example, Daily 12:00 AM). Select + Add to add additional schedules. The page displays the computed Next Execution date and time in your selected time zone. Shown for Scheduled Summary and Instant + Summary.

Regardless of the notification timing option selected, you can also turn on Also send a notification on successful runs to receive a confirmation after each successful monitored policy or rule run. Successful runs notified this way do not create an incident.

Naming and advanced settings

Enter an Incident Policy Name (required) and, optionally, select + Add description to add a description.

Expand Advanced Settings to configure:

Incident Life Cycle

• Default assignee (optional): The user incidents are assigned to by default. Defaults to Unassigned.
• Auto-resolve incident when conditions clear: When enabled, set the number of consecutive passing evaluations required before an open incident is automatically resolved. The incident closes as soon as the next evaluation passes the trigger threshold.

Runbook

Add one or more runbooks, each with a Runbook name, a URL, and an optional Description. Runbook links are shown in incident details and in notifications. Select + Add Runbook to add additional runbooks.

Select Create Policy to save the Incident Policy.

What's Next

• Configure notification channels used by Incident Policies. See [Notification Integrations].
• Review incidents raised by your policies on the Incidents tab.