This internal table is available starting with oblogservice 1.1.1.
Purpose
Records resource event alerts for a cluster. When the resource usage of the cluster exceeds or falls below a specific threshold, LogService generates a resource event alert, indicating that some resources are overloaded or underloaded. Cluster administrators can poll this internal table to subscribe to resource event alerts, which helps determine whether the cluster needs to be scaled up or down.
Columns
Field name |
Type |
Nullable |
Description |
|---|---|---|---|
resource_type |
u32 | NO | The resource type (primary key). Valid values: 0 = IOPS; 1 = BPS (bandwidth). |
az |
String | NO | The name of the availability zone (AZ) (primary key), for example, az1, az2, or ALL_AZ (indicates all AZs). |
ln |
u64 | NO | The log node (LN) ID (primary key). A value of 0 indicates an event at the AZ level (not the node level). |
event |
u32 | NO | The event type. Valid values: 0 (OverUtilized) and 1 (UnderUtilized). |
utilization_rate |
u32 | NO | The resource utilization (percentage, 0-100). For example, 80 indicates 80%. |
comment |
String | NO | The event description, which describes the cause or details of the event. |
start_time |
i64 | NO | The event start time (in microseconds). |
Sample query
Query resource event alerts for a cluster:
ls_ctrl -H <HOST_ADDRESS> show meta --table-name t_resource_event