Azure Automation and Azure Virtual Machines are pleased to announce the release of a new feature allowing you to configure Virtual Machine (VM) alerts to run Automation runbooks. This new capability allows you to automatically perform standard remediations in response to VM alerts, like restarting or stopping the VM.
Previously, during VM alert rule creation you were able to specify an Automation webhook to a runbook in order to run the runbook whenever the alert triggered. However, this required you to do the work of creating the runbook, creating the webhook for the runbook, and then copying and pasting the webhook during alert rule creation. With this new release, the process is much easier because you can directly choose a runbook from a list during alert rule creation, and you can choose an Automation account which will run the runbook or easily create an account.
It is easy to set up an Azure VM alert and configure an Automation runbook to run whenever the alert triggers – see this article for step-by-step instructions. For this initial release you can choose from three runbooks that the service provides – Restart VM, Stop VM, or Remove VM (the ability to choose other runbooks or one of your own runbooks is coming soon). Example scenarios include restarting a VM when the memory usage exceeds some threshold due to an application on the VM with a memory leak, or stopping a VM when the CPU user time has been below 1% for past hour and is not in use.
As part of the process of associating a runbook with an alert we will create an Automation account automatically for you if you don't have one yet. Also, we also will create an Azure Run As account in the Automation account to simplify the authenticate aspect of management of Azure resources with runbooks.
After you configure a runbook for an alert, you can disable it without removing the configuration. This allows you to keep the alert running and test the alert rules and then later re-enable the runbook.
Summary
When you configure an alert on an Azure VM, you now have the ability to easily configure an Automation runbook to automatically perform remediation action when the alert triggers. In this release, you can choose from runbooks to restart, stop, or delete a VM depending on your alert scenario. This is just the beginning of enabling scenarios where you control the actions (notification, troubleshooting, remediation) that will be taken automatically when an alert triggers.
If you have feedback on this feature or ideas on scenarios you would like see enabled then put your comments on Azure User Voice and vote up the ideas of others.
Just getting started with Azure Automation? Learn about the service here, and follow Azure Automation on Twitter.