Information Points

[CommunityUser]

Hi Evereyone,

I am new to this forum, and looking for some help in configuring Information Points.

We are using Introscope as our JVM monitoring tool. But we are migrating to AppDynamics now. We have AppDynamics agent set up on all our servers. And now we are migrating the custom configurations.

We have lot of classes/methods that are defined in Introscope pbd files. These methods are not any service end points, or entry points just some random methods that call the database, external vendor calls, and so  on. 

We figured out that we can add these methods as Information Points in AppDynamics. We have this application deployed on 2 servers. But now the question is that we dont have any mechanism to distinguish between the metrics collected from server A vs Server B. Information point doesnt tell me the avg response time based on a particular server. 

Please suggest.

Thanks,

Payal

[Jean-Baptiste_M]

No indeed you cannot get that per tier or node. However, you said your
definition s are "random" calls to DB, external vendors and so on...
So in that case I'd suggest to use custom exit points instead of IP (if
these exit points not already detected ootb), this is meant for your use
cases.
JB

[CommunityUser]

Hi,

Thanks for your response. I am trying to use custom exit points and I have configured a few. Two questions:

1) Where can I see the defined custom exit points in metric browser? I am not able to find that.

2) Why is "exit points" defined in "Backend detection" section? As I understand, it could be anything - any method call - then why it is is "Backend detection" section?

Thanks,

Payal

[Jean-Baptiste_M]

1/ In Backends

2/ Backends are detected through exit point rules. A backend is something that is called by your code but which is "remote" (ie not inside your JVM

The idea here is you said those """"random""" classes makes calls to backend (DB, external services and so on), therefore a backend (Even locally)

Not sure to be clear enough 😉