---
title: "Get started with Alert Management"
description: "Create an email monitoring integration, create a CPU alert, attach it to a running Instance, and know how a notification looks."
content_type: "get-started"
product: "alerts"
platform: "TIR"
url: "/tir/alerts/get-started/"
---

# Get started with Alert Management

Set up an alert that emails you when an Instance in your project passes a CPU threshold. You need
a TIR project with a running Instance in it. To create one, see
[Get started with TIR Instances](/tir/tir-instance/get-started/).

## Create a monitoring integration

1. In TIR, open your project. In the side menu, under **PLATFORM SERVICES**, select **Alert
   Management**.
2. Select **Manage Monitoring Integrations**, then **Create Monitoring Integration**.
3. Keep or replace the generated **Integration name**.
4. Leave **Contact point type** on **Mail**.
5. In **Members**, pick yourself and each person who needs the email.
6. Select **Create integration**.

Notifications go to the email address of each member you
picked.

## Create an alert

1. Select **Manage Alerts** to return to the alert list, then **Create Alert**.
2. Keep or replace the generated **Alert name**.
3. In **Service for alert**, choose **Instance**.
4. In **Monitoring Integration**, choose the integration you created.
5. Under **Alert Trigger**, set **Alert trigger type** to **CPU Utilization**, **Severity** to
   **Warning**, **Operator** to **Greater Than**, and **Threshold value** to 85.
6. Select **Create**.

The alert is created. Next, attach it to a resource so it
starts watching.

## Attach it to an Instance

1. Open your running Instance and select its **Alert** tab.
2. Under **Alerts**, open **Select Alerts**, choose your alert, and select **Attach**.

The alert is now watching the Instance, with the status **OK**.

## What happens next

As soon as CPU Utilization goes above 85 percent, the status in the table changes to **Firing**. If
it is still above 85 percent after five minutes, TIR emails the members of the integration. When it
drops back within the threshold, the status returns to **OK** and TIR sends a second email that says
the alert is resolved.

To send alerts somewhere else, add [Slack](/tir/alerts/how-to/send-alerts-to-slack/) or a
[webhook](/tir/alerts/how-to/create-a-monitoring-integration/#create-a-webhook-integration) as a
destination. To alert on other metrics, see the [metrics each service
offers](/tir/alerts/concepts/alert-metrics-and-conditions/).