This page provides an overview of alerting policies. For detailed how-to information, see the following:
You can set up alerting policies in the Stackdriver Monitoring Console to determine when your applications are not operating normally. When external events trigger conditions in one of your alerting policies, Stackdriver Monitoring displays an incident and sends notifications to people or third-party notification services. Responders can acknowledge receipt of the notification.
A Stackdriver account can have no more than 500 alerting policies.
Alerting policy components
You configure your alerting policies using the Stackdriver Monitoring Console. A policy belongs to an individual Stackdriver account, so be sure to select the account where you want your policies. Each policy has a name, one or more trigger conditions, optional notifications, and optional documentation.
To see and manage your policies, select Alerting > Policies Overview in the Stackdriver Monitoring Console, or click the following button:
To create a new alerting policy, click Add Policy in the Alerting Policies page, select Alerting > Create a Policy in the Stackdriver Monitoring Console, or click the following button:
Conditions determine when an alerting policy opens an incident. You create the conditions when you create your alerting policy. Conditions can reference the following:
- Predefined metrics from individual VM instances, third party applications, and other monitored resources such as databases and load balancers.
- Custom metrics and logs-based metrics that you create yourself.
- Health checks on processes, VM instances, or URL endpoints.
The specific types of conditions include Metric Threshold, Metric Rate of Change, Metric Absence, Group Aggregate Threshold, Process Health, and Uptime Checks.
An alerting policy can have no more than 6 conditions.
When the conditions in an alerting policy are triggered, Stackdriver Monitoring opens an incident, which is recorded and displayed in the Stackdriver Monitoring Console. To see your open incidents, select Alerting > Incidents, or click the following button:
To contact specific people or services when an incident is created, configure one or more notifications when you create your alerting policy. In addition to email and SMS notifications, you can connect with third-party applications including as PagerDuty, HipChat, Campfire, and others. For more information, see Notification Options.
An alerting policy can have no more than 16 notification channels.