Merge pull request #222602 from AbbyMSFT/patch-20

Update dynamic thresholds troubleshooting question

Author: prmerger-automator[bot]

articles/azure-monitor/alerts/alerts-dynamic-thresholds.md

@@ -105,12 +105,11 @@ If the behavior of a metric changed recently, the changes won't necessarily be r
 
 To identify weekly seasonality, the Dynamic Thresholds model requires at least three weeks of historical data. When enough historical data is available, any weekly seasonality that exists in the metric data is identified and the model is adjusted accordingly.
 
-## Dynamic Thresholds shows a negative lower bound for a metric even though the metric always has positive values
+## Dynamic Thresholds is showing values that are not within the range of expected values
 
-When a metric exhibits large fluctuation, Dynamic Thresholds builds a wider model around the metric values. This action can result in the lower border being below zero. Specifically, this scenario can happen when:
+When a metric exhibits large fluctuation, Dynamic Thresholds builds a wider model around the metric values. This model can result in a lower border below zero when the metric only has positive values, or in an upper border above 100% when the metric can't exceed 100%. This scenario can happen when:
 
 - The sensitivity is set to low.
-- The median values are close to zero.
 - The metric exhibits an irregular behavior with high variance, which appears as spikes or dips in the data.
 
 When the lower bound has a negative value, it's plausible for the metric to reach a zero value given the metric's irregular behavior. Consider choosing a higher sensitivity or a larger **Aggregation granularity (Period)** to make the model less sensitive. Or, use the **Ignore data before** option to exclude a recent irregularity from the historical data used to build the model.
@@ -306,4 +305,4 @@ Use the following information to interpret the chart:
 - **Red dot with a black circle**: Shows the first metric value out of the allowed range. This value fires a metric alert and puts it in an active state.
 - **Red dots**: Indicate other measured values outside of the allowed range. They won't fire more metric alerts, but the alert stays in the active state.
 - **Red area**: Shows the time when the metric value was outside of the allowed range. The alert remains in the active state as long as subsequent measured values are out of the allowed range, but no new alerts are fired.
-- **End of red area**: When the blue line is back inside the allowed values, the red area stops and the measured value line turns blue. The status of the metric alert fired at the time of the red dot with black outline is set to resolved.
+- **End of red area**: When the blue line is back inside the allowed values, the red area stops and the measured value line turns blue. The status of the metric alert fired at the time of the red dot with black outline is set to resolved.