data_storage_warning_tolerance_time specifies the tolerance period after which the state of an abnormal data disk is set to WARNING.
Note
Starting from OceanBase Database V4.0.0, the default value is changed from 30s to 5s, and the value range is changed from [10s, 300s] to [1s, 300s].
| Attribute | Description |
|---|---|
| Type | Time |
| Default value | 5s |
| Value range | [1s, 300s] |
| Effective upon OBServer node restart | No |
When an unexpected I/O failure occurs, the I/O manager of OceanBase Database starts a detection thread to detect the data disk status. If the data disk status fails to become normal within the period specified by this parameter, the detection thread sets the data disk status to WARNING. This state triggers events such as a leader switchover to respond to business requests.
Example
obclient> ALTER SYSTEM SET data_storage_warning_tolerance_time = 20s;