Track device CPU load within a moving time frame. See periods where CPU load is abnormally peaked (load bottlenecks, service attacks, or other service incidents) or abnormally slow (server dropped from load balancer configuration, kernel panic) and more.

Configure a CPU utilization threshold:

  • Name. Used in the Threshold library and the title on the Alert Center Dashboard.
  • Notification Policy. (Optional) Select the notification policy to apply to this threshold. The policy initiates notifications when an item is outside the configured threshold limits.
  • Threshold Check Interval. Enter a time interval for Alert Center to check the WhatsUp Gold database for items that are out of the threshold limits.

    Automatically resolve items no longer out of threshold. Select this option if you want Alert Center to automatically resolve items when they return to the value within the threshold limit.

注: Notification policies are optional for most thresholds. If you do not select a notification policy, no notifications are generated for the threshold, but a dashboard report listing the out of threshold items still appears on the Alert Center Home page.

Add Condition Rules:

  • Report. Select CPU utilization threshold monitor method, device or CPU option.
    • Device-wide combined CPU utilization. Select this option to calculate the threshold based on the average-combined CPU load evaluated as a single device. Therefore, if the average CPU performance of a quad-core CPU (an average of all CPU performance combined) exceeds the threshold, then an alert is triggered.
    • Individual CPU utilization. Select this option to calculate the threshold based on individual CPU load for the selected device. Therefore, if one of the CPUs on a quad-core CPU exceeds the threshold, then an alert is triggered.
    • CPU Utilization (average). Choose a trend (Exceeds, Falls-below) and a value for average CPU load in percent.
  • Duration. Specify a time range or "rolling interval" for calculating an average to compare to CPU Utilization threshold.
    ヒント: Notification policies are optional for most thresholds. Out of threshold items still appear on the Alert Center dashboard.

注: Configure the threshold check interval for a longer time than the sampling interval for thresholds relating to trends, such as percent utilization. Configure it for a time the same as (or similar to) the sampling interval when configuring a threshold for a health check.

ヒント: Avoid setting the threshold check interval to a very short time, as this can degrade system performance. In general, setting the threshold check interval to less than five minutes is not advised.

適用されたデバイスの選択

  • [Apply to All Devices (すべてのデバイスに適用)]。デフォルトでは、しきい値は適用可能なすべてのホストを監視します。

    [デバイスを除外]。除外リストを作成または適用する場合に選択します。

  • [このしきい値を特定のデバイスに適用] をクリックして、しきい値を適用するホストを選択します。
    注: 使用率などの傾向に関連するしきい値には、しきい値チェック間隔にサンプリング間隔よりも長い時間を設定してください。正常性チェックのしきい値を設定する場合は、サンプリング間隔と同様の時間に設定してください。