We frequently upgrade our Amazon Elastic Compute Cloud (EC2) fleet, with many patches and upgrades being applied to instances transparently. However, some updates require a short instance reboot in order to apply these updates. This page provides information on the reboot process.
The recent release of our scheduled events functionality provides greater visibility into the timing of these reboots. In addition to added visibility, in most cases you can use the scheduled events to manage reboots on your own schedule if you want to reboot before the scheduled update window. You can easily view any upcoming scheduled events for your instances in the AWS Management Console or using the API tools or command line. Reboots such as these should be infrequent, but may be necessary from time to time to apply upgrades that strengthen our security, reliability and operational performance.
There are two kinds of reboots that can be required as part of Amazon EC2 scheduled maintenance – instance reboots and system reboots. Instance reboots are reboots of your virtual instance, and are equivalent to an operating system reboot. System reboots require reboots of the underlying physical server hosting an instance. If you do not take any action, the impact on your instance is the same in both cases – during your scheduled maintenance window your instance will experience a reboot that in most cases takes a few minutes.
You also have the option to manage these reboots yourself at any time prior to the scheduled maintenance window. When you manage a reboot yourself, your instance will receive the upgrade when you reboot and your scheduled maintenance window will be cancelled (note that scheduled events can sometimes take up to 1 hour to refresh once a reboot has been completed).
Q: How do I manage my own Instance Reboots
Managing your own instance reboots is very straightforward. Also note that in place of the AWS Management Console process listed below, you can also reboot your instance from the operating system of your instance.
- Log into the AWS Management Console and go to the Scheduled Events area on the EC2 tab to see a listing of any of your instances that are scheduled for instance reboots (you can find this under the Event Type column). Note: this step is optional, as you can also find these instances in Step 2 below.
- Go to the Instance page and find those instances with scheduled instance reboots (instances with scheduled events will have stopwatch icons next to the instance IDs.)
- Select the check box next to the instance you want to manually reboot, click Instance Actions, and then select Reboot from the drop-down list. Please be sure not to shutdown or terminate.
- Once your instance has rebooted (this normally takes a few minutes), verify that your application is working. At this point, your instance should no longer have a scheduled event associated with it or the scheduled event will display “[Completed]”. Note that it can occasionally take up to 1 hour for this status to refresh
For details on how to view and manage your scheduled events using the API and command line tools, please see Monitoring the Status of Your Instances in the Amazon EC2 user guide.
Q: How do I manage my own System Reboots
It is not possible for you to perform your own system reboots. It is recommended that you let AWS perform system reboots for you. When AWS performs a system reboot, none of your configurations such as DNS name or IP address will change, and data on the local instance store will remain intact.
If it is critical for you to manage the upgrade of an instance that needs a system reboot, you will need to re-launch your instance using the process below, at which point your instance should no longer have a scheduled event associated with it.
EBS-backed AMIs: If you are running an EBS-backed AMI, you can stop and then restart your instance in order to easily re-launch it. This will cause the loss of any data you have saved on the local instance store of the instance, thus you should back up that data before stopping your instance if you wish to keep that data. In addition, the public DNS name and internal IP address of your instance will change (except when running in Amazon VPC). You will also need to re-associate any Elastic IP for your instance. For a detailed list of additional configuration settings that change, see Monitoring the Status of Your Instances in the Amazon EC2 User Guide.
Instance store-backed AMIs: If you are running an instance store-backed AMI, you will need to re-bundle your AMI and launch a new instance. This will cause you to lose any data you have saved on the local instance store of the instance, and will change your internal IP (except when running in Amazon VPC).
For more details on how to complete both of these tasks, see Monitoring the Status of Your Instances and Bundling Amazon EC2 instance store-backed Linux/UNIX AMIs in the Amazon EC2 User Guide.
Q: How long does a reboot take?
A reboot typically completes within a few minutes. The exact time varies depending on the instance configuration.
Q: Does a reboot change any configuration settings for my instance?
No, neither instance reboots nor system reboots have any effect on your configuration settings. Rebooted instances retain all configuration data including:
- Public DNS name
- The private IP address
- Any data stored on the instance store (e.g. data not on the EBS volume)
- An associated elastic IP address (if one is used)
Q: How can I check if a scheduled reboot is completed?
You can use the AWS Management Console, the command line, or the API. If your scheduled event no longer appears or displays as “[Completed]”, the reboot has been completed.
Q: Will new instance launches need to be rebooted?
No, if you launch a new instances it will already have the updates and will not require a reboot.
Q: Why are you doing these reboots?
These reboots are needed to apply upgrades that strengthen our security, reliability and operational performance.