How to quickly enable a set of smart alarms to monitor your Autonomous Databases

October 24, 2023 | 4 minute read
Keith Laker
Senior Principal Product Manager
Text Size 100%:

The "autonomous" nature of the Autonomous Database means that many of the regular database management operations are fully automated or can be fully automated if the relevant tickboxes are ticked - such as auto-scaling for compute resources and auto-scaling for storage. Over time, the state of a database can evolve as new data is added by applications or data integration tools, data structures get updated, more users log on to run queries, the pattern of those queries changes, etc.

Therefore, for many enterprise customers running mission-critical applications, there are lots of situations where fleet and cloud administrators want to be notified about specific types of events related to the operational status of their mission-critical Autonomous Databases. 

When you are responsible for a fleet of Autonomous Database instances, it can be helpful to define your additional alarms that trigger notifications that can send emails, SMS messages, and Slack messages so you can quickly respond to specific situations or create automated workflows to deal with those situations. Fortunately, the Autonomous Database is a genuine cloud-native service, which means it is tightly integrated into the typical cloud telemetry services (unlike some other so-called cloud-native databases that are black boxes, completely sealed off from the supporting cloud infrastructure).

How do you create new alarms?

Each Autonomous Database comes with a library of built-in metrics that are visible on the OCI console. If you scroll down the page to the resources section, there is a metrics view that displays eight graphs covering CPU utilization, storage utilization, sessions, running statements, queued statements, etc. Each chart has a default statistic (sum, mean, max, min, etc), which can be changed via the menu.



The options button on the top right corner of each graph contains a menu that allows you to create a new alarm based on the metric within the chart:



It is a relatively quick process, and the pop-up form guides you through the various steps. The following video walks you through the process of creating a new alarm:


This process is quick and easy if you only manage one Autonomous Database. If you have a large estate of mission-critical Autonomous Databases to manage, you need a smarter way to create the required alarms.

Is there a built-in library of the most useful alarms?

The short answer is: Yes!

As mentioned, the Autonomous Database is tightly integrated into the OCI telemetry services. This means that Autonomous Database administrators can use the Database Management console, which has a library of built-in alarms that can be deployed in a few clicks. The video below walks you through this process:

In summary, you must enable the “Database Management” service under “Associated services”. The enablement process takes a couple of minutes, and then you are ready to switch over to the Database Management console.


What does the library of alarms monitor?

On the Alarm Definitions page, you can very easily add a whole suite of useful alarms that provide the following types of monitoring:

  • Critical - Alarm fires when the Mean of CpuUtilization is greater than the threshold value of 85.
  • Critical - Alarm fires when the Mean of FailedLogons is greater than the threshold value of 30.
  • Warning - Alarm fires when the Mean of FailedLogons is greater than the threshold value of 20.
  • Warning - Alarm fires when the Mean of Sessions is greater than the threshold value of 30.
  • Critical - Alarm fires when the Mean of StorageUtilization is greater than the threshold value of 85.
  • Warning - Alarm fires when the Mean of StorageUtilization is greater than the threshold value of 75.

These alarms can be modified to use different thresholds using the edit feature if the pre-configured thresholds don't match your precise requirements.


If you have a fleet of mission-critical Autonomous Databases, then please save yourself some time and effort and enable the "Database Management" feature and make use of the built-in library of alarms. You can set up and assign your notification processes to these pre-built alarms to send emails, SMS messages, slack messages, etc. This is the fastest way to jump-start your monitoring framework for all your mission-critical Autonomous Database instances.


Keith Laker

Senior Principal Product Manager

Product Manager for Autonomous Database and Analytic SQL with extensive experience in data warehouse and business intelligence projects. Worked in various roles, including post-sales consultancy, customer support, and product management at locations across Europe and US.


Previous Post

Autonomous Database Learning Lounge

Marcos Arancibia | 15 min read

Next Post

Oracle Autonomous Data Warehouse is really fast. Try it for yourself.

Hermann Baer | 2 min read
Everything you need to know about data warehousing with the world's leading cloud solution provider