Comparing Metrics Between Applications Manager and Azure Portal

Overview

This guide covers the general navigation steps applicable to all Azure monitors whose data is collected via the Azure APIs.

General Navigation Steps

1. Log in to the Azure Portal based on your cloud environment:
   • Azure Global: https://portal.azure.com/
   • Azure Government Cloud: https://portal.azure.us/
   • Azure China Cloud: https://portal.azure.cn/
2. Search for the respective Azure resource type in the top search bar (e.g., App Services, ExpressRoute circuits, Load balancers).
3. Select the specific resource instance being monitored in Applications Manager.

For Performance Metrics

4. In the left navigation pane, go to Monitoring → Metrics.
   
5. Select the respective metric name and aggregation type based on the metric description in Applications Manager.
   • Metric: Select the metric from the dropdown that corresponds to the metric shown in Applications Manager. Refer to the metric tooltip or the Help documentation for the mapping (e.g., Data In in Azure Portal corresponds to Incoming Traffic in Applications Manager).
   • Aggregation: Select the aggregation type as mentioned in the metric tooltip or the Help documentation.
     • Example: Consider a metric described in Applications Manager as: "The total amount of network traffic transmitted per second between the poll interval (in MB/s)."
       • Aggregation - Sum: The keyword "total" in the description maps to the Sum aggregation in the Azure Portal.
         • "Between the poll interval" - indicates that the value aggregates all data points collected within the specified time range (i.e., the poll interval).
         • "At the time of polling" - indicates that the value is the most recent data point from the specified time range.
       • Unit - MB/s (Derived): The unit MB/s indicates that this is a derived metric. The raw value returned by the Azure API is in Bytes, which Applications Manager converts to MB/s.
   
   Notes: 

   • If the exact aggregation type is unclear from the tooltip or Help documentation, try the available options (e.g., Average, Sum, Count). The goal at this stage is simply to confirm whether data is populated in the Azure Portal. You can also add multiple metrics to the chart simultaneously for comparison.
   • Some metrics in Applications Manager are derived metrics, their values are converted from the raw unit returned by the Azure APIs (e.g., Bytes to MB/s, BitsPerSecond to Mb/s). When comparing such metrics with the Azure Portal, a difference in values is expected due to unit conversion and is not an error.
6. Set the Time Range and Time grain to cover the period of the reported discrepancy or missing data.
   • Time Range: Use an extended window such as Last 1 hour, Last 3 hours and so on. Avoid matching the exact polling interval - a wider range ensures data is visible even if there are gaps or delays.
   • Time grain: Set to the lowest available granularity, such as 1 minute, to confirm whether data exists within the selected range.
   • Example: If the polling interval is 5 minutes, set the Time Range to Last 1 hour and Time grain to 1 minute.
   

   Once the time range and time grain are set, check whether the chart shows data. The two states are illustrated below:

   Metric with Data	Metric with No Data
   Chart shows metric values plotted over the selected time range.	No data here — the metric has no data in Azure Portal.

   

   Note: If the chart shows no data in the Azure Portal, the data does not exist in Azure and will therefore not be collected in Applications Manager. If the resource has active usage but no data appears in the Azure Portal for the metric, please contact Azure Support.
7. If the metric requires filtering (e.g., by dimension or property), click Add filter, then select the appropriate Property and Value.
   • Data for some metrics in Applications Manager is collected with dimensions. To check the metric value, a filter must be applied in the Azure Portal.
   • Example: For Azure Load Balancer's SNAT Connection Count metric, click Add filter → set Property as Connection State → set Value as Failed, Pending, or Successful to match the specific metric in Applications Manager.
   
   

For Configuration Metrics

4. In the left navigation pane, click Overview.
5. On the Overview page, click JSON view (available in the top-right area of the Overview pane).
6. In the JSON view, search for the respective parameter name corresponding to the configuration metric and verify its value.

Note: Configuration metrics are not available under Monitoring → Metrics in the Azure Portal.

Still Facing Issues?

If the values still do not match after following the steps above, or if you are unable to locate the metric in the Azure Portal, follow the steps below to capture diagnostic information and share it with support for faster resolution.

Step 1 - Enable Debug Logging

1. In Applications Manager, enable the following logging options:
   1. Go to Settings → Logging → General Logging → Select Print all logs.
   2. Go to Settings → Performance Polling → Cloud tab → Enable Log API Responses. This captures the raw response received from the Azure APIs and helps identify whether the discrepancy originates from the API itself.

Step 2 - Trigger Data Collection and Wait

1. Trigger a manual poll for the affected monitor: navigate to the monitor's details page and click the Poll Now button (available in the monitor action menu or at the top of the monitor page).
2. Wait for data collection to complete fully before proceeding. Do not trigger another poll or restart the application during this time.

Step 3 - Collect and Share the SIF

1. Once data collection is complete, generate and download the Support Information File (SIF) from Applications Manager:
   • Go to Settings → Support → Create SIF.
   • The SIF bundles all relevant logs, configuration details, and API responses needed for analysis.

Step 4 - Share the Following

Share the following details:

• The monitor's name and monitor type.
• A screenshot of the metric value shown in Applications Manager.
• A screenshot of the equivalent value in the Azure Portal.
• A screenshot of the Microsoft Azure Parent monitor's Edit Monitor page (navigate to the monitor's details page → Monitor Actions → click Edit Monitor to capture the configurations).
• A screenshot of the Alarm History for the affected monitor, showing the Health and Availability history (navigate to the monitor's details page → Monitor Actions → Alarm History → Health History & Availability History tab).
• The generated SIF file.

Additionally, for Performance Metrics, include:

• Screenshots of the metric chart containing Time Range & Aggregation in the Azure Portal - both with and without filters applied (if applicable).
• The polling interval configured in Applications Manager for the monitor.

Note: For Performance metrics, use an extended Time Range (e.g., Last 1 hour or Last 3 hours) with a low Time grain (e.g., 1 minute) when capturing the Azure Portal screenshot.