Calculate metrics based on APIM API prodcut

Venkatesh Korrapati 20

Hi Team,

While creating shared dashboard for each API in APIM. I see issues with product based grouping.
To calculate health(uptime, error rate) , Performance metrics (response time, latency ) and Success/failures rate percentage metrics for each API in APIM .

using re-emit policy I am unable to fetch above metrics. Could you please suggest how to achieve this using product Id based filtering.

Thanks

Venkat

Contact:+1(913-951-6781).

Vinod Pittala 155 Reputation points Microsoft Vendor

2025-02-01T00:28:23.4833333+00:00
Hello Venkatesh Korrapati,

Welcome to Microsoft Q&A Forum, thank you for posting your query here!

Before creating a shared dashboard, ensure that your APIM instance is sending metrics and logs to Azure Monitor or Application Insights.

First, enable Application Insights for your APIM if it is not already enabled. This will allow you to collect telemetry data for API calls, errors, response times, etc.

Please refer to the document which can guide you in enabling the insights.

https://learn.microsoft.com/en-us/azure/api-management/api-management-howto-app-insights?WT.mc_id=Portal-fx&tabs=rest

Then, enable Diagnostic Settings for your APIM to send logs and metrics to your monitoring solution.

Please refer to the below MS article:

https://learn.microsoft.com/en-us/azure/api-management/diagnostic-logs-reference

below are the example logs that needs to be enabled on Diagnostic Settings.

GatewayLogs, RequestLogs, PerformanceLogs.

Once these logs send to the destination, you can monitor them by querying. (FYR only)

Below is the example query in Application Insights (Health, Latency, Error Rate).

requests | where url contains "your-api-name" | where customDimensions["productId"] == "your-product-id" | summarize avg(duration), count() by resultCode, bin(timestamp, 1h) | extend successRate = count_ / (count_ + failedCount_)

-->productId can be a custom dimension you’ve added during the Re-Emit policy or can be part of the request/response headers.

Query in Azure Monitor (for API Metrics by Product).

AzureDiagnostics | where ResourceType == "APPLICATIONGATEWAY" // Or APIM depending on logs | where ProductId_s == "your-product-id" | summarize avg(ResponseTime_s), count(), avg(Latency_s) by bin(TimeGenerated, 1h)

This query returns the average duration, count of requests, and calculates the success rate based on the status code.

-->To use the Re-emit policy in Azure API Management, you can insert this policy into the API’s inbound or outbound processing pipeline to log custom metrics. You would typically use this for advanced telemetry and capturing API-specific performance.

<inbound> <base /> <re-emit> <event name="apim-api-usage"> <metric name="responseTime" value="@(context.ResponseTime)" /> <metric name="errorRate" value="@(context.Faults.Count)" /> <metric name="successRate" value="@(context.Response.StatusCode >= 200 && context.Response.StatusCode < 300 ? 1 : 0)" /> </event> </re-emit> </inbound>

Once you have your queries set up and verified, you can add them to a shared Azure Dashboard.

Here’s how to create and configure a dashboard for each API:

Create a new custom dashboard.

Add Tiles for Metrics:

Add a tile to the dashboard for each API metric.

For example, add a Metrics Chart or Log Query to display health metrics, response time, and error rates.

Select Azure Monitor or Application Insights as the source for the tile.

Use the relevant metric or Kusto query you created earlier.

For each tile, ensure you apply filters like API Name, Product ID, or other relevant identifiers for specific APIs.

Organize Tiles.

By following these steps, you will have created a shared dashboard that allows you to monitor the health, performance, and success/failure rates of each API in Azure API Management (APIM).

Let me know if you have any queries.

Thank you.
Rahul Podila 1,725 Reputation points Microsoft Vendor

2025-02-03T02:15:10.8433333+00:00

Hi @Venkatesh Korrapati

Just want to check if the above Comment worked for you or else please let us know if any help, we are always here to help whenever you need us.

If the comment is helpful, please click "Upvote it"

Thankyou
Venkatesh Korrapati 20 Reputation points

2025-02-03T17:59:14.1633333+00:00
Hi @vinod pittala

Thank you for detailed information.

I am already following all these steps. Below are my concerns.

Seems <re-emit> policy doesn't exists and it gives lot of errors.

Step:1 added emit policy in APIM global policy inbound section

<emit-metric name="Request"> <dimension name="Product ID" /> </emit-metric>

Step2: but it always shows the value as "1",

FYI same application insights is attached to APIM, app gateway, function app etc;,

I want to achieve below operational metrics for each API with product ID grouping and no where I found proper documentation on this.

-Health Metrics:

Uptime: The percentage of time the API is available.

Error Rates: The frequency of errors occurring in the API.

-Performance Metrics:

Response Time: The time taken to respond to API requests.

Latency: The delay before a transfer of data begins following an API request.

Throughput

Success/Failure Metrics:

Success Rate: The percentage of successful API requests.

Failure Rate: The percentage of failed API requests.

When I run below query, failedCount_ doesn't exists in "requests" table.

requests | where url contains "your-api-name" | where customDimensions["productId"] == "your-product-id" | summarize avg(duration), count() by resultCode, bin(timestamp, 1h) | extend successRate = count_ / (count_ + failedCount_)

Thanks

Venkat
Madugula Jahnavi 0 Reputation points Microsoft Vendor

2025-02-04T05:09:27.1266667+00:00

Hello Venkatesh Korrapati, Can you brief about what is the exact requirement and what is the exact issue facing now with the above query!

1 answer

Madugula Jahnavi 0 Reputation points Microsoft Vendor

2025-02-06T13:58:09.31+00:00

Hello @Venkatesh Korrapati,
Thank you for reaching out to Microsoft Q&A!

To calculate or send custom metrics to app insight loggers, there is no policy called re-emit available in APIM policy references as you also mentioned. There is an existed policy called emit-metric policy which sends custom metrics to application insights or azure monitor and the policy format you used is accurate. So now, adding an emit-policy in the global APIM should work but here there are some other conditions you need to check.

Firstly, check if the logger diagnostics has been enabled to "True" to receive the API logs based on the given filtering.

And also, there are some limitations when sending metrics through emit policy like You can configure a maximum of 10 custom dimensions per policy.

Reference MS Doc for all the limitations: https://learn.microsoft.com/en-us/azure/api-management/api-management-howto-app-insights?tabs=rest#limits-for-custom-metrics Check here and validate your metrics accordingly.

Also, make sure that you set "with dimensions" field under linked Application insights >> Usage and estimation costs >> custom metrics (preview) >> select "with dimensions" and click on ok. It should be enabled before sending metric dimensions as shown below.

Mentioning above, you can also refer this https://learn.microsoft.com/en-us/azure/api-management/api-management-howto-app-insights?tabs=rest#emit-custom-metrics MS Doc on how to integrate the API with Application insights clearly.
Please sign in to rate this answer.

0 comments No comments
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

Calculate metrics based on APIM API prodcut

1 answer

Your answer