Showing results for 
Show  only  | Search instead for 
Did you mean: 
Not a customer? Start a free trial

Click the Start a free trial link to start a 15-day SaaS trial of our product and join our community as a trial user. If you are an existing customer do not start a free trial.

AppDynamics customers and established members should click the sign in button to authenticate.

Dynamic Languages (Node.JS, Python, PHP, C/C++, Webserver Agent)

End-to-end tracing for application return results multiple times


End-to-end tracing for application return results multiple times

My POC application has two node.JS applications. The 1st one is based on Microsoft Botframework and the 2nd one just exposes APIs for first one to consume.


I found that when I send a request to the 1st app (Chatbot), the application first returned a debug event with HTTP status code 200 very quickly. Actually the Chatbot app is still doing work and it calls the 2nd application to get the information it needs, the Chatbot app will return another result to the caller after a few seconds, e.g.

[10:56:47] -> POST 202 [message] I want to reserve jade dragon 
[10:56:48] <- POST 200 Reply[event] Debug Event 
[10:56:53] <- POST 200 Reply[message] There is no seat available on 4-June


In AppDynamics, I can see both endpoints (the 1st one is the call to Chatbot and the 2nd one is to the API), but they are not connected. The 1st endpoint reported with very quick response time (~10 ms). It looks to me that the system assumes the call is finished and ignore the subsequent HTTP POST event.


Is there any way to link these two applications? The fact is the Debug Event returned is meaningless from application performance management perspective - it was just an echo that the bot framework has received something from the requester.

By replying you agree to the Terms and Conditions of the AppDynamics Community.
End-to-end tracing for application return results multiple times
AppDynamics Team

HI Ricky,


Are you using our SaaS environment? If so, could your share your account name (not access key). I'll be able to better understand what you're experiencing by looking at the data directly.




Kyle Furlong, Technical Lead (C++ and Dynamic Languages)

Found something helpful? Click the Accept as Solution button to help others find answers faster.
Liked something? Click the Thumbs Up button.

Yes I am on SaaS with the same account posting this message.

Basically I gave up. I am evaluating multiple APM solutions in parallel and I got similar results. I understand that I need to write custom code if we want to measure end-to-end response time in this case.