The records of auto scaling actions can be viewed in Scaling history. Grading of auto scaling events is supported, and event alarm policies are set based on the event level. For more information about event levels, see Cluster Events. For more information about event alarm configuration, see Alarm Configurations. Custom Scaling Records
Supports filtering the scaling records by execution time range and searching by policy name.
By default, untriggered rules are not displayed. To view all scaling records, you can disable Show Only Triggered Rules.
Sorts by Execution time, and displays Execution time, Policy name, Scaling type, and Execution status. You can click Details in the Operation column to view details.
There are four auto scaling execution statuses
Not triggered: The condition for auto scaling (scale-out/scale-in) is not met; the rule is not triggered.
Retrying: If auto scaling cannot execute at the specified time due to various reasons, the system will retry within the set time range.
In progress: The auto scaling activity is being executed.
Successful: According to the scaling rule, all target numbers of nodes are successfully added to or removed from the cluster.
Partially successful: According to the scaling rule, some nodes are successfully added to or removed from the cluster, while others fail due to disk quota management or CVM inventory.
Failed: According to the scaling rule, no nodes were added to or removed from the cluster.
Scaling Group Scaling Records
You can filter scaling records by execution time period and search for them by policy name.
By default, untriggered rules are not displayed. To view all scaling records, you can disable Show Only Triggered Rules.
The list displays the execution time, scaling group name, policy name, scaling type, and execution status. You can sort by execution time and filter by scaling type and execution status. You can click Details under the operation type to view detailed information, or click Analyze to view policy execution details, primarily used for troubleshooting failure reasons.
The execution status of auto scaling includes the following 6 types:
Not triggered: The condition for auto scaling (scale-out/scale-in) is not met; the rule is not triggered.
Retrying: If auto scaling cannot execute at the specified time due to various reasons, the system will retry within the set time range.
In progress: The auto scaling activity is being executed.
Successful: According to the scaling rule, all target numbers of nodes are successfully added to or removed from the cluster.
Partially successful: According to the scaling rule, some nodes are successfully added to or removed from the cluster, while others fail due to disk quota management or CVM inventory.
Failed: According to the scaling rule, no nodes were added to or removed from the cluster.
Managed scaling records
Supports filtering the scaling records by Execution Time or Scaling Type.
The list displays the execution time, scaling type, model specifications, quantity, execution status, and reason. You can sort by execution time and filter by scaling type and execution status.
The execution status of managed scaling includes the following 2 types:
Successful: The target node is added to or removed from the cluster based on the cluster load.
Failed: The target node failed to be added to the cluster based on the cluster load due to resource insufficiency. We recommend you modify the preset resource specification.
Model Specification: The target model specification and type after the rule is triggered for a scale-out or scale-in.
Quantity: The number of added or removed nodes in various specifications after the operation.