The latest cloud infrastructure announcements, technical solutions, and enterprise cloud insights.

  • October 18, 2018

Improved Availability of Your Instances with Customer-Managed VM Maintenance

Sanjay Pillai
Principal Product Manager

We are excited to announce customer-managed virtual machine (VM) maintenance, a major step in Oracle Cloud Infrastructure’s ongoing effort to improve the availability of your VM instances. You can now easily reboot your instances and avoid scheduled downtime for planned infrastructure maintenance.

What is customer managed VM maintenance?

Today when an underlying infrastructure component needs to undergo maintenance, we notify you in advance of the planned maintenance downtime. To avoid this planned downtime, you can opt to terminate and re-redeploy your instances prior to the planned maintenance. With the introduction of customer managed VM maintenance, we give you another option. Instead of terminating and re-deploying your instance manually, you can now reboot your instance from the Console, API or CLI. This new experience makes it easy for you to control your instance downtime during the notification period.

The reboot or restart of a VM instance during the notification period is different from a normal reboot. The reboot or stop/start workflow stops your instance on the existing VM host which needs maintenance and starts it on a healthy VM host. Customer managed VM maintenance makes it easier for you to avoid the planned maintenance downtime. If you choose not to reboot during the notification period, then Oracle Cloud Infrastructure will reboot your instance for you before we proceed with the planned infrastructure maintenance.

How do I get started?

Getting started is easy. When there is a maintenance event, Oracle Cloud Infrastructure will notify you via email. You can identify the affected VMs in the Console by checking the Maintenance Reboot field (or check the timeRebootMaintenanceDue property using the API/CLI) - which will show the date and time after which the infrastructure maintenance will occur. The instance reboot will occur within a 24 hour period following the specified time. Both the Instance list view and the Instance details view (below) will display the Maintenance Reboot field.

For Standard VM instances with a boot volume, additional iSCSI block volume attachments, and a single VNIC, you can proceed to reboot or stop and start the instance. If you have non-iSCSI (paravirtualized or emulated) block volume attachments or secondary VNICs, you must detach them first before rebooting or restarting your instance.

When you reboot or stop and start the instance, it is migrated to a different physical VM host - while preserving all the instance configuration properties, including ephemeral and public IP addresses. When the Reboot Maintenance field is blank, the instance is no longer impacted by the maintenance event.

Finding affected instances

To make it easier to find and act on your instances, you can search for the instances that have been set to reboot in your tenancy by using the Advanced Search and choosing the Query for all instances which have an upcoming scheduled maintenance reboot sample query.

Customer-managed VM maintenance is currently supported on Standard VM instances running Linux OS. It supports instances launched from Oracle Cloud Infrastructure images and images imported from external sources. It is offered in all regions at no extra cost.  

To learn more about customer-managed VM maintenance on Oracle Cloud Infrastructure, see the Best Practices for Your Compute Instances. For more information about the Oracle Cloud Infrastructure Compute service, see the Oracle Cloud Infrastructure Getting Started guideCompute service overview, and FAQ.

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.Captcha