Docs

Elastic Compute Service

Event monitoring

Event type overview

This topic summarizes the types, statuses, names of, and recommended actions for currently supported events.

Events of cloud resources

Events triggered by faults, operations, or status changes during the operation of cloud resources

Scheduled operations events

Event type	Event name	Event	Event description and impact	Recommended action on the tenant side
SystemMaintenance.Redeploy	Redeployment for instances of system maintenance	SystemMaintenance.Redeploy:Inquiring SystemMaintenance.Redeploy:Executing SystemMaintenance.Redeploy:Succeeded SystemMaintenance.Redeploy:Failed SystemMaintenance.Redeploy:Rejected SystemMaintenance.Redeploy:Canceled	VolcanoEngine detected that the host machine or dedicated host of the ECS instance is in a suboptimal state, such as CPU overheating, power failure, or fan malfunction. This may cause downtime risks.	For instances without mounted local volumes Manual authorization You need to select the maintenance time slot and give permission for the redeployment of the instance onto a new host. For detailed actions, refer to "Instance redeployment due to system maintenance" event. Automatic authorization You need to create a scheduled maintenance window in advance to authorize VolcanoEngine to redeploy the instance on a new host. tip If VolcanoEngine has not obtained your permission for O&M operations, the event will be automatically canceled when there are changes to the host machine (such as type changes, power-on, and restart). This is an automatic O&M process. The automatic cancellation (automatic O&M) is triggered only for instances of General-purpose, Compute-optimized, Memory-optimized, and High Frequency families. For details, refer to the instance type list.
InfrastructureUpgrade.Redeploy	Instances redeployed due to infrastructure upgrade	InfrastructureUpgrade.Redeploy:Inquiring InfrastructureUpgrade.Redeploy:Executing InfrastructureUpgrade.Redeploy:Succeeded InfrastructureUpgrade.Redeploy:Failed InfrastructureUpgrade.Redeploy:Canceled InfrastructureUpgrade.Redeploy:Rejected	As business develops and the health status of host machine of ECS instances changes, VolcanoEngine will upgrade the infrastructure that is outdated or has poor performance.	Manual authorization You need to select the maintenance time slot and authorize the redeployment of the instance onto a new host. For specific steps, refer to Querying and responding to the "Redeployment for infrastructure upgrade instances" events. Automatic authorization You need to create a scheduled maintenance window in advance to authorize VolcanoEngine to redeploy the instance on a new host. tip If VolcanoEngine has not obtained your permission for O&M operations, the event will be automatically canceled when there are changes to the host machine (such as type changes, power-on, and restart). This is an automatic O&M process. The automatic cancellation (automatic O&M) is triggered only for instances of General-purpose, Compute-optimized, Memory-optimized, and High Frequency families. For details, refer to the instance type list.

Unexpected operation and maintenance events

Event type	Event name	Event	Event description and impact	Recommended action on the tenant side
SystemFailure.Redeploy	The instance is redeployed due to system failures	SystemFailure.Redeploy:Failed SystemFailure.Redeploy:Succeeded SystemFailure.Redeploy:Inquiring SystemFailure.Redeploy:Executing SystemFailure.Redeploy:Canceled	It is detected that the ECS instance shut down due to failures in the host machine and dedicated host. The instance needs to be redeployed to a new host to restore normal operations. Otherwise, the instance and its services stop running.	If the triggered event is "SystemFailure.Redeploy:Executing", no actions are required, because the instance will be automatically deployed on another host by default. You can also manually adjust the automatic maintenance mode. For detailed actions, refer to the Querying and responding to the "Redeployment due to instances of system failure" event. If the "SystemFailure.Redeploy:Inquiring" event is triggered, give permission to the redeployment of the instance.
SystemFailure.Reboot	System failure instance restart tip Only instances of Big Data, Local SSD, and HPC-optimized instance type families can trigger this event. For details, refer to the instance type list.	SystemFailure.Reboot:Executing SystemFailure.Reboot:Succeeded SystemFailure.Reboot:Failed	A system failure on the host machine of the ECS instance is detected. The instance will be restarted to restore the running status.	If the restart fails, your permission is required to redeploy the instance. tip By default, the instance automatically restarts. If the restart fails, the instance will be automatically redeployed. You can also manually adjust the automatic maintenance mode. For detailed actions, refer to the Querying and responding to the "System failure instance restart" event.
SystemFailure.Stop	The instance is stopped due to system failures	SystemFailure.Stop:Executing SystemFailure.Stop:Succeeded SystemFailure.Stop:Failed	It is detected that the ECS instance shuts down due to failures in the host machine.	No action is required.
FileSystemReadOnly.Change	File system became read-only	FileSystemReadOnly.Change:Executing FileSystemReadOnly.Change:Succeeded	The file system of the mounted disk of the ECS instance becomes read-only. At this time, the disk can only be read but not written. Reasons can be disk hardware failure or system file corruption.	No action is required.
InstanceFailure	An operating system error instance exception	InstanceFailure:Succeeded	A kernel panic has been detected in the operating system of an ECS instance during its runtime.	We recommend that you check whether the instance exceptions exist. If no instance exception exists, no action is required on your part. If any instance exception exists, you need to restart the instance manually to restore the instance normal operation. tip For more rapid troubleshooting of kernel crashes in the instance, we recommend that you enable the Kdump service for the operating system: Enabling Kdump service in Linux operating system Enabling kernel-mode dump in the Windows operating system
GpuRiskDetected	GPU risk detected tip Only GPU Accelerated and HPC-optimized instance type families can trigger this event. For details, refer to the instance type list.	GpuRiskDetected.Succeeded	An exception has been detected in your running GPU instance, or an exception is predicted. You need to restart the instance to restore the normal instance operations. tip If the instance is not restarted, it may lead to downtime or performance exceptions, which could impact your business continuity.	You need to restart the instance manually.
DiskErrorDetected	Hard disk exception tip Only instances of Big Data, Local SSD, EBM mounted with local volumes, and Heterogeneous Compute-optimized (GPU and HPC) instance type families can trigger this event. For details, refer to the instance type list.	DiskErrorDetected:Inquiring DiskErrorDetected:Succeded DiskErrorDetected:Canceled	Exceptions are detected in the hard disk of the Local SSD instances and cause exceptions for read and write operations. You need to give permission to the redeployment of the instance to a new host or replacement of the hard disk. Otherwise, the instances stop running.	You need to give permission for the redeployment of the instance or replacement of the volume with exceptions. For details, refer to the Local volume event O&M.
DiskError.Redeploy	Instance redeployment due to hard disk failure tip After you grant permission on instance redeployment, this event is generated.	DiskError.Redeploy:Executing DiskError.Redeploy:Succeeded DiskError.Redeploy:Canceled	Due to hardware or software failures in the hard disk of Local SSD instances, VolcanoEngine is deploying the instances with exceptions onto new hosts to restore the normal operation of the instances.	No action is required.
DiskError.ReplaceDisk	Hard disk replaced due to exception tip After you grant permission on the replacement of the faulty hard disk, this event is generated.	DiskError.ReplaceDisk:Executing DiskError.ReplaceDisk:Succeeded DiskError.ReplaceDisk:Failed DiskError.ReplaceDisk:Canceled DiskError.ReplaceDisk:Pending	Due to hardware or software failures in the hard disk of Local SSD instances, VolcanoEngine is replacing the faulty hard disk with a new one for instance normal operations.	No action is required.
InstanceOOM	Instance memory OOM	InstanceOOM:Succeeded	The OOM is detected in ECS instance memory.	No action is required.
InstanceFileSystemFailure.StopAndRepair	Downtime repair due to file system exception tip Currently, only Windows instances can trigger this event.	InstanceFileSystemFailure.StopAndRepair:Inquiring InstanceFileSystemFailure.StopAndRepair:Canceled InstanceFileSystemFailure.StopAndRepair:Execting InstanceFileSystemFailure.StopAndRepair:Failed InstanceFileSystemFailure.StopAndRepair:Succeed InstanceFileSystemFailure.StopAndRepair:Rejected	It is detected that file system exceptions occur due to the damaged system volume of the instance. This may cause startup issues or connection failures.	You need to select a maintenance time slot and give permission to VolcanoEngine for shutdown repair operations. For details, refer to Authorized shutdown repair. tip After the instance enters the repair status, the instance status is MAINTENING, and you can not start the instance, replace the system volume, initialize the system volume, or change the specification.
InstanceDiskStalled	Seriously impaired disk performance for instance	InstanceDiskStalled.Executing InstanceDiskStalled.Succeeded	On the mounted volume of the ECS instance, IO congestion occurs. This may severely impact volume performance and thereby effect normal read/write operations.	We recommend that you isolate read and write operations to the volume at the application level or temporarily remove the instance from the load balancer list.

Fee warning events

Event type	Event name	Event	Event description and impact	Recommended action on the tenant side
AccountUnbalanced.Stop	Instance stopped due to account arrears	AccountUnbalanced.Stop:Executing AccountUnbalanced.Stop:Succeeded	The pay-as-you-go instance will stop in 1 hour because the associated account is in arrears.	We recommend you to recharge your account in time for renewal.
AccountUnbalanced.Delete	Instance released due to account arrears	AccountUnbalanced.Delete:Executing AccountUnbalanced.Delete:Succeeded	The pay-as-you-go instance will be released in 3 hours because the associated account is in arrears.	We recommend you to recharge your account in time for renewal.

Status change event

Event type	Event name	Event	Event description and impact	Recommended action on the tenant side
CreateInstance	Instance created	CreateInstance:Executing CreateInstance:Succeeded CreateInstance:Failed	When you create an ECS instance, this system event is triggered.	No action is required. tip To improve the instance creation success rate, it is recommended: Before creating an instance, call the DescribeAvailableResource interface to query the instance supply status in the destination availability zone and ensure sufficient private IP address numbers in the virtual switch. If the destination available resources are insufficient, you can use the Auto Scaling (AS) capability to expand the resource pool.
RunInstance	Run instances	RunInstance:Executing RunInstance:Succeeded RunInstance:Failed	When you start an ECS instance, this system event is triggered.	No action is required.
StopInstance	Instance stopped	StopInstance:Executing StopInstance:Succeeded StopInstance:Failed	When you stop an ECS instance, this event is generated.	No action is required.
DeleteInstance	Instance deletion	DeleteInstance:Executing DeleteInstance:Succeeded DeleteInstance:Failed	When you delete an ECS instance, this system event is triggered.	No action is required.
SpotInstanceInterruption.Delete	Spot instance was interrupted and instance was released	SpotInstanceInterruption.Delete:Scheduled SpotInstanceInterruption.Delete:Executing SpotInstanceInterruption.Delete:Succeeded	When resource inventory is insufficient or the spot market price exceeds your bid price, the system reclaims your spot instances. This system event is triggered 5 minutes before the reclamation.	No action is required.
InstanceChargeType.Change	Instance billing method changed	InstanceChargeType.Change:Executing InstanceChargeType.Change:Succeeded InstanceChargeType.Change:Failed	When you change the billing method of an ECS instance, this system event is triggered.	No action is required.
InstanceConfiguration.Change	Instance configuration changed	InstanceConfiguration.Change:Executing InstanceConfiguration.Change:Succeeded InstanceConfiguration.Change:Failed	When you upgrade or downgrade an ECS instance, this system event is triggered.	No action is required.
RebootInstance	Restart instances	RebootInstance:Executing RebootInstance:Succeeded RebootInstance:Failed	When you restart an ECS instance, this system event is triggered.	No action is required.
DeploymentSet.Modify	Instance deployment set modified	DeploymentSet.Modify.Executing DeploymentSet.Modify.Succeeded DeploymentSet.Modify.Failed	When you change the deployment set of an ECS instance, this system event is triggered.	No action is required.
ServerMigrationTask	Server migration task	ServerMigrationTask.Executing ServerMigrationTask.Pending ServerMigrationTask.Recovered ServerMigrationTask.Succeeded ServerMigrationTask.Failed	When you start to use the SMC for migrating source server data, this system event is triggered.	No action is required.
ServerMigration.FirstSync	Server migrated and full data synced	ServerMigration.FirstSync.Succeeded ServerMigration.FirstSync.Failed	When you use the SMC for full migration of source server data, this system event is triggered.	No action is required.
ServerMigration.AdditionalSync	Server migration and incremental data sync	ServerMigration.AdditionalSync.Succeeded ServerMigration.AdditionalSync.Failed	When you use the SMC for incremental migration of source server data, this system event is triggered.	No action is required.

Task events

Events triggered by status changes of cloud resource tasks

Task status class event

Event type	Event name	Event	Event description and impact	Recommended action on the tenant side
ElasticScheduledInstance.Create	An ESI order created	ElasticScheduledInstance.Create:Failed ElasticScheduledInstance.Create:Executing ElasticScheduledInstance.Create:Succeeded	When you create an ESI order, this system event is triggered.	No action is required.
ElasticScheduledInstance.Cancel	An ESI order cancelled	ElasticScheduledInstance.Cancel:Failed ElasticScheduledInstance.Cancel:Executing ElasticScheduledInstance.Cancel:Succeeded	When you cancel an ESI order, this system event is triggered.	No action is required.
ElasticScheduledInstance.Deliver	An ESI order delivered	ElasticScheduleInstance.Deliver:Scheduled ElasticScheduleInstance.Deliver:Canceled ElasticScheduledInstance.Deliver:Failed ElasticScheduledInstance.Deliver:Executing ElasticScheduledInstance.Deliver:Succeeded ElasticScheduledInstance.Deliver:Pending	When an ESI order is delivered, this system event is triggered.	No action is required.
ElasticScheduledInstance.Invalid	An ESI order became invalid	ElasticScheduledInstance.Invalid:Succeeded	When an ESI order expires, this system event is triggered.	No action is required.
ElasticScheduledInstance.AutoReleaseAt	Elastic Scheduled Instances (ESI) Automatic Release	ElasticScheduledInstance.AutoRelease:Scheduled ElasticScheduledInstance.AutoRelease:Executing ElasticScheduledInstance.AutoRelease:Succeeded	This system event is generated when an ESI order is automatically released.	No action is required.