Does a manual stop/start clear an AWS_LIGHTSAIL_INSTANCE_MAINTENANCE_SCHEDULED event, and should the Health Dashboard still show "Upcoming"?

Question

I received an AWS Health notification of scheduled maintenance for a Lightsail instance in ap-southeast-2, with a 2-hour window later this month.
Following the guidance in the notification and in the Knowledge Center article "How do I manage a scheduled maintenance notification in Lightsail?", I did a full Stop, waited for the Stopped state, then Start - not a Reboot.
I have good evidence the instance did move to new hardware. It is a Xen-based instance, so host-derived CPU values are visible in the guest, and I captured them immediately before the stop and again after the start:
- CPU microcode revision changed (0xb000040 -> 0xd000433)
- TSC calibration changed (2300.081 MHz -> 2299.998 MHz)
- /proc/cpuinfo flags checksum changed
- CPU model, instance ID, AZ, MAC and private IP all unchanged
Despite that, the AWS Health Dashboard still shows the event as "Upcoming" with 1 affected entity, well over an hour later.
The Knowledge Center article says the stop/start "changes the status of the scheduled maintenance event from Active to Complete" and that it "can take up to 1 hour to reflect the status change". But I have also seen an answer from an AWS expert on re:Post saying the Health Dashboard simply does not pick up a manual stop/start and will keep showing the event as upcoming.
Questions:
1. Which behaviour is correct for Lightsail today - should the event clear, or is "Upcoming" expected to persist until the window passes?
2. 2. Is there any supported way to verify that a Lightsail instance's scheduled maintenance has been satisfied, other than the Health Dashboard?
3. 3. Given the hardware evidence above, is it safe to assume no stop or reboot will occur during the scheduled window?
4. This is a production system, so I need to know whether to plan for an outage. Thanks.

re:Post Agent · Answer

I need to clarify an important limitation: the sources I have access to cover Amazon EC2 scheduled maintenance events, not Amazon Lightsail. While both services run virtual instances, they have different maintenance notification systems and behaviors.

For EC2 instances, a manual stop and start does migrate the instance to new hardware and clears scheduled maintenance events. The instance status typically updates within about an hour to show the event as completed. However, Lightsail uses a different event notification system (AWS_LIGHTSAIL_INSTANCE_MAINTENANCE_SCHEDULED) that operates separately from the EC2 scheduled events framework.

Regarding your specific questions about Lightsail:

1. I cannot definitively answer whether the Health Dashboard should clear the event for Lightsail instances, as this is service-specific behavior not covered in my available sources.

2. For verification methods specific to Lightsail maintenance events, I don't have authoritative information in my sources.

3. While your hardware evidence (CPU microcode, TSC calibration changes) strongly suggests the instance migrated to new physical hardware, I cannot confirm whether this guarantees the scheduled maintenance is satisfied for Lightsail specifically.

4. Given this is a production system and you need certainty about potential outages, I recommend adding correspondence to your existing support case through the AWS Support Center (https://console.aws.amazon.com/support/) with the hardware evidence you've collected. The support team can verify whether your instance's maintenance obligation has been satisfied and provide definitive guidance on whether to expect any disruption during the scheduled window. You might also consider reviewing your AWS Support plan options if faster response times would benefit your production operations.  
**Sources**  
[Manage Amazon EC2 instances scheduled to stop or retire - Amazon Elastic Compute Cloud](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/schedevents_actions_retire.html)  
[Manage Amazon EC2 instances scheduled for reboot - Amazon Elastic Compute Cloud](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/schedevents_actions_reboot.html)  
[Manage Amazon EC2 instances scheduled for maintenance - Amazon Elastic Compute Cloud](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/schedevents_actions_maintenance.html)

Does a manual stop/start clear an AWS_LIGHTSAIL_INSTANCE_MAINTENANCE_SCHEDULED event, and should the Health Dashboard still show "Upcoming"?

Add your answer

Relevant content