Setting Default Thresholds

This section describes how to set alarm thresholds and alarm clearance thresholds for performance metrics of specific objects in a storage device. If a metric value is above (or below) the alarm threshold, DeviceManager sends an alarm to notify you of checking and handling the device fault. If a metric value is above (or below) the alarm clearance threshold, DeviceManager clears the alarm.

Procedure

  1. Choose Settings > Monitoring Settings > Monitoring Threshold > Default Threshold.
  2. After selecting an object from the Object Type list, you can view all performance metrics of the object.

    • Default thresholds can be configured for the following object types:
      • Converged services: Cluster - File Service, Cluster - Object Service, Cluster - HDFS Service, Remote Replication Pair, Port, and Disk.
      • Block service: Remote Replication Pair, Initiator, Storage Pool, Port, Volume, and Disk.
    • To select Cluster - File Service, Cluster - Object Service, or Cluster - HDFS Service, enable the corresponding service first.

  3. Click Modify to modify the threshold settings in the table. Table 1 describes the related parameters.

    Table 1 Threshold parameters

    Parameter

    Description

    Metric Name

    Name of a specific object metric.

    [Example]

    Read Bandwidth (KB/s)

    Threshold Type

    Threshold type of a specific object metric. If the metric value is above or below the threshold, the system generates an alarm. The options are > and <.

    [Default value]

    >

    [Example]

    >

    Alarm Threshold

    When the value of a specific object metric reaches the value of this parameter for multiple consecutive times (multiple consecutive times refers to the value of Trigger Condition (Times)), the system generates an alarm.

    [Default value]

    --

    [Example]

    60

    Alarm Clearance Threshold

    When the value of a specific object metric reaches the value of this parameter for multiple consecutive times (multiple consecutive times refers to the value of Trigger Condition (Times)), the system clears the generated alarm.

    [Default value]

    --

    [Example]

    50

    Trigger Condition (Times)

    When the number of times that the value of a specific object metric reaches Alarm Threshold reaches the value of this parameter, the system generates a threshold alarm.

    When the number of times that the value of a specific object metric reaches Alarm Clearance Threshold reaches the value of this parameter, the system clears the generated threshold alarm.

    [Default value]

    6

    [Example]

    6

    NOTE:

    You can determine the alarm sampling period based on the interval between two consecutive performance data samplings in the performance monitoring chart of the corresponding object performance metric.

    • If the interval is less than 1 minute, the alarm sampling period is 10 seconds.
    • If the interval is greater than or equal to 1 minute, the alarm sampling period is 1 minute.

    Severity

    Severity of a threshold alarm. Possible values are Critical, Major, and Warning.

    [Default value]

    Warning

    [Example]

    Warning

    Threshold Status

    Switch for controlling whether a threshold is enabled. The preceding parameters can be set only after this switch is turned on.

    [Default value]

    Disabled

    [Example]

    Enabled

    NOTE:

    If Threshold Status for a reported alarm is Disabled, the alarm will be cleared automatically.

    Take the parameter values in the Threshold parameters table as an example. When Threshold Status for Read Bandwidth (KB/s) is Enabled, an alarm of the Warning severity will be generated if the read bandwidth (KB/s) exceeds 60 KB/s for 6 consecutive times, and will be cleared if the read bandwidth (KB/s) falls below 50 KB/s for 6 consecutive times.

  4. (Optional) Click Reset to clear the settings and restore the default settings.
  5. Click Save.