Datadog

This integration allows:

• Users in Monte Carlo to click a button to convert an alert into an Incident in Datadog, or link an existing Incident. Multiple Datadog Incidents may be linked or created from a single alert in Monte Carlo.
• Monte Carlo to automatically send notifications directly to Datadog to create Incidents.
• Status and ownership updates from Monte Carlo alerts are synced to the corresponding Incident in Datadog.

Integration setup

An Account Owner must go to Settings > Integrations. In the Notifications and Collaboration section, click Create and select Datadog, and complete the form.

Required information

Input:

• Display name: The name that will appear for this integration within Monte Carlo
• Site: The Datadog site (region) for your organization (e.g., datadoghq.com, us5.datadoghq.com, datadoghq.eu)
• API Key: An organization-level key for API access that identifies which Datadog account the request belongs to
• Application Key: A key linked to the user (or service account) who created it that grants permissions and authorizes actions based on that user's roles and defined scopes

Creating Datadog API and Application Keys

To create the required keys in Datadog:

1. API Key: Go to Organization Settings > API Keys and create a new API key
2. Application Key: Go to Organization Settings > Application Keys and create a new application key

Required scopes

The Application Key must have the following scopes:

• incident_write (required): Allows creating and updating incidents
• user_access_read (required): Allows reading user details to support owner/commander sync

Optional scopes

• incident_read: Required only when linking existing incidents

Syncing Status and Ownership from Monte Carlo to Datadog

When enabled, Monte Carlo will automatically sync alert status and ownership changes to the corresponding Datadog incident. This is a one-way sync (Monte Carlo → Datadog):

Status Mapping:

• Monte Carlo statuses "No status", "Acknowledged", "Work in progress" → Datadog "Active"
• Monte Carlo statuses "Fixed", "Expected", "No action needed" → Datadog "Resolved"

Priority Mapping:

Monte Carlo alerts can have an assigned priority (P1–P5), depending on the monitor definition that triggered the alert. When creating a Datadog incident from an alert, Monte Carlo maps the alert’s priority to the corresponding Datadog incident severity as follows:

• No priority → UNKNOWN
• P1 → SEV1
• P2 → SEV2
• P3 → SEV3
• P4 → SEV4
• P5 → SEV5

Owner Sync:

When the owner changes in Monte Carlo, the corresponding Datadog incident commander is updated if a Datadog user with the same email is found.

Creating Datadog incidents from Monte Carlo alerts

After the integration is set up, you can create Datadog incidents either manually from an alert page or automatically via Audiences.

• Manual (from an alert page)

  1. In the top-right toolbar of the alert page, click Linked issues (chain icon) and select Add Datadog incident.
  2. In the side panel, choose Create new or Link existing. Linking an existing incident requires the Application Key to include the incident_read scope.
  3. For a new incident, review or edit the Title, Description, and Incident Commander. Title and Description are pre-populated with alert details and a link back to Monte Carlo.
  4. Click Create. You'll see a confirmation with a one-click link to view the incident in Datadog. You can repeat this to link multiple Datadog incidents to the same alert.

• Automated (via Audiences)

  1. Go to Settings > Notifications and audiences and click Create audience, or edit an existing audience.
  2. Add a Datadog recipient channel and select your Datadog integration.
  3. Associate the audience with your monitors or other notification rules. Any alert sent to that audience will create a Datadog incident.

Interacting with Datadog incidents in Monte Carlo

Once a Datadog incident is created or linked to an alert, you can:

• In the top-right toolbar of the alert page, open Linked issues (chain icon). Under the Datadog section, each linked incident appears by its number (for example, #200).
• Click the incident to open it in Datadog.
• To remove a link, click the Unlink icon next to the incident number in the Linked issues menu. This only removes the association from the alert in Monte Carlo; the incident remains in Datadog.