It would be great to see raw body request on analytics. Good for monitoring restful api calls (for example json payloads)102 votes
We’d like to hear a bit more about specifically what you have in mind with regard to this suggestion. Please reply to this suggestion with as much detail about what you’d like.
The average response time metric is not very useful if it is counting signalr persistent connections. I can't filter out these requests in the default dashboard graphs and it's a pain to filter it out in every single graph I create. There needs to be a way to ignore certain patterns from being tracked.44 votes
When an exception happens and the details are captured on Application insight, it would be great if POST parameters were available as well.
This feature can be turned off for certain pages via configuration r even posting an extra parameter like Ignore Post Parameters.
This will make sure sensitive application data like login information is not captured.
This feature can only work when a http POST ends up with a 500 error to assist debugging application, to make sure huge data is not captured and never used.24 votes
There are natural PII concerns in auto-collecting POST parameters, and it’s something, as Dan mentioned, you can do yourself with telemetry initializer. That said, please keep upvoting so we can see how many folks are impacted by not having a more streamlined way of collecting POST parameters.
-AppInsights Product Management
App Insights drops the last octet of an IP. The only way to collect full IP is to do it manually or mess with telemetry initializers. Not a big deal, but IPs are not so sensitive and I don't think these should be obfuscated at all. At least give us an option (ideally in the portal) to collect the full IP.16 votes
Last octet is dropped by AI by design, since it represents PII (personally identifiable information). As you mentioned you can do this through the SDK or enrichment (i.e. customer can proactively collect PII, but not out of box). Can you describe a scenario where the full IP is required to triage/diagnose a production issue?
There should be a priority level that indicates the relative priority of the data.
The priority level of a tracked item should be possible to set on an item by item basis, or by more general declarative rules.
High priority data should be retained for longer time and care should be taken to make sure no high priority data is lost anywhere.
Example of use case:
You might want to mark events when a user change/request a new password longer than a metric of time taken for a storage request.10 votes
Having different retention policies for different telemetry items is a very interesting feature request. Enabling this is fundamentally non-trivial, so please keep upvoting and including the use cases where this is necessary. Please keep in mind that you can today ensure retention longer than 90d by using continuous export to archive subsets of the telemetry to external storages/data warehouses of your choice.
-AppInsights Product Management
We'd like to use Application Insights with our native (Win32) server application written in C++, as well as our Win32 desktop application (C++).
Without C++ support we can't build a full stack solution based on Application Insights.9 votes
Add support for VueJs that supports the entire Analytics feature set. There is an open source one currently, but its pretty basic.6 votes
Can you clarify what Analytics feature set you are referring to?
Morgan from the AppInsights team
It would be nice if it is clarified which character encoding is supported by Application Insights and SDKs.
We may want to send telemetry with application logs via LogBack and AI Appender but the log may have multi-bytes local language characters.5 votes
You used to be able to get Screen Resolution by creating a new Session chart and Grouping By "Screen Resolution". This option is no longer there! How can find this info now so it can help me make decisions (please don't let it be send custom events... i need to look at my current historic data)?4 votes
I like the idea of monitoring WCF Services with Application Insights and I'm sucessfully using the related SDK Lab already in production for a while. Please make the WCF Lab SDK offical! In addition, it would be nice to be able to log the RAW SOAP Messages as well.4 votes
Need url referral on pageViews, and requests. How to keep track where the visitor come from?4 votes
I want to collect my telemetry data for my cdn and be able to tell
- my overall cdn metrics for my app integrated in my solution
- if I have a gain when enabling cdn
- where I should leverage cdn4 votes
It would be great if there would be a SDK client which allows of querying data (metrics, events etc) instead of having to make REST API calls.
Even if it might be trivial to make REST calls, from a developer perspective it would be a big improvement to just be able to call a method without having to deal with auth or looking up the call you need to make.3 votes
IE11 mistakenly gets categories as IE7.3 votes
I confirm, don't get in Page View in most of the iPhone.3 votes
When exploring custom performance counters in Application Insights , the sample interval is by default 1 second, should there be a way to customize this in ApplicationInsights.config?3 votes
Per Sergey’s request:
Please reactivate GitHub issue and tell us what is your specific scenario why you need the collection interval to be shorter giving other characteristics of Application Insights backend. Assumptions is that you need more than just collection interval change and we need to know what.
The article at  provides guidance on using tags to annotate requests so that we can separate traffic coming from separate environments and such. However, adding tags right now requires explicitly writing an IContextInitializer.
It would be great if specifying the tags would be available as an option directly from the ApplicationInsights.config file, either by directly providing an element for including the tag data, or by providing a built-in IContextInitializer that read it from the application configuration.3 votes
When I tried to track exceptions thrown by WCF services using telemetry.TrackException of the Microsoft.ApplicationInsights library then the exception was tracked but a exception was also thrown for System.ArgumentException: 'Value does not fall within the expected range.' The stack trace showed the following:
at System.Runtime.InteropServices.Marshal.ThrowExceptionForHRInternal(Int32 errorCode, IntPtr errorInfo)
at System.Web.Hosting.IIS7WorkerRequest.GetServerVariableInternal(String name)
The exception did not mention which argument and what value was affected. So its very hard to find the root cause of the issue and arrive at a solution. Screenshot of the code and exception has been attached2 votes
Add adaptive sampling for Node.js SDKs
"Adaptive sampling automatically adjusts the volume of telemetry sent from the SDK in your ASP.NET app." - https://docs.microsoft.com/en-us/azure/application-insights/app-insights-sampling2 votes
Current data submission through AI client requires a Key.
Storing a key in an application is not desirable. It will be a good idea if the AAD-authenticated applications (Service Principals) could be used instead of passing the key.
See Azure Key Vault integration with Azure Active Directory for an example of this works2 votes
- Don't see your idea?