Setting up alarms for PostgreSQL DBaaS instance
This section describes the steps to create alarms for PostgreSQL flexible server using Azure and AWS portals.
The alert triggers when the value of a specified metric crosses the assigned threshold. The alert triggers when the condition is met and when that condition is no longer being met.
Create an alert rule on a metrics from the Azure portal
- In the Azure portal, select the Azure Database for PostgreSQL server you want to monitor.
- Under the Monitoring section of the sidebar, select Alerts.
- Select Create alert rule or +.
- On the Create an alert rule page, under the Condition tab select Add condition.
- Select a metric from the list of signals to be alerted on from the Select a signal page.
- Configure the alert logic including the Condition (for example, Greater than), Threshold (for example, 85 percent), Time Aggregation, Period of time the metric rule must be satisfied before the alert triggers (for example, over the last 30 minutes), and Frequency.
- Select Next: Actions >.
- Under the Actions section, select Create action group to create a new group to receive notifications on the alert.
- Fill in the Basics form with a name, display name, subscription, and resource group. Select Next: Notifications >.
- Configure an Email/SMS message/Push/Voice action type by providing the details for all the required fields and then click OK.
- (Optional) Select Next: Actions > to add actions based on the alerts.
- Select Review+Create to review the information and create.
- Provide the alert details. and select Next/Review+Create.
If the alert triggers, an email would be sent to the provided email id in the notification section.
For more information, refer to the PostgreSQL section of the Azure documentation.
The following procedure is an example for creating an alarm that notifies the users if they run out of storage space on their RDS instance.
Create an alert rule on a metrics from the AWS portal
- Open the CloudWatch console.
- In the navigation pane, navigate to Alarms > All Alarms.
- Select Create alarm > on the Select metric page > search for FreeStorageSpace metric > select RDS > select Per-Database metrics.
For the instance that you want to monitor, select the DB instance identifier FreeStorageSpace metric.
- Click on Select metric.
- In the Specify metric and conditions step, provide the required details.
- In the Configure actions step, provide the required details.
After the details are provided and you click on Create topic, a confirmation email would be sent to the provided email address.
Open the email notification that you received from AWS Notifications, and then click on Confirm subscription.
- Return to the Configure actions page in the CloudWatch console and click Next.
- Enter a name and description for your alarm, and then click Next.
- Review the preview of your metric, and then select Create alarm.
Note:
It is a best practice to create a second, critical alarm for a lower threshold. For example, set your first alarm for 25 GB, and the second critical alarm to 10 GB.
For more information on creating alarms for other critical metrics, refer to the Amazon RDS User Guide.
To configure event notifications for Amazon RDS
- Open the CloudWatch console.
- In the search bar, search for Simple Notification Services and select Topics from the left pane.
- Select Create Topic > Standard.
Enter the required details and click Create.
- Select Create Subscription > Email from the list of protocols.
Enter the email address on which you want to receive notifications.
Note:
Confirm the created subscription.
- Navigate to Amazon RDS console > Event subscriptions > Create event subscription.
Enter the Name, select the ARN for the SNS topic, and select Instances as the Source type.
Select specific instances and select your instance.
- Navigate to Select specific event categories > select Maintenance > Create.
For more information, refer to the RDS maintenance section of the Amazon Knowledge Center.