Showing results for 
Show  only  | Search instead for 
Did you mean: 

Can we treat unmonitored internal endpoints like an external call?


Can we treat un-monitored internal endpoints like an external call? DT SaaS lets you easily do this for external Web services calls and this enables us to alert and track counts etc. DT kind of walked you through this easy enough with requests to external, but when we try to do that with requests to unmonitored internal hosts it doesn’t give us the option to monitor it separately – it wants us to add DT SaaS to it. We get this but since that isn’t easy (we don't always have a say in the tool picked by our downstream internal partners) - can we get DT SaaS to show us how to seperate it out like it’s an external call even though it’s inside?


Dynatrace Helper
Dynatrace Helper

I'd like to see this as well.

In some cases these 'Requests to Unmonitored Hosts' are on devices or hosts that we can't put the OneAgent (Mainframe, Data Power, etc.). I'd like to be able to simply select specific requests and give them their own Service name so they can track certain requests uniquely, pin them to the Dashboard, etc.

Dynatrace Champion
Dynatrace Champion

This is actually possible, but a bit hidden. you can create a custom device via the API that you give the respective IP:port. Once that custom device exists the requests to it will result in a new service that "sits" on top of that custom device.

Can you delete a custom device once one is made? I successfully made a device, but want to delete it now as it was just a test. I see deleting a custom metric in the link you provided, but not deleting the device.

As far as I remember, custom device vanishes from smartscape and database after 72 hours since last data has been pushed to this entity. I think it applies for all entities in general.

Certified Dynatrace Master | Alanata a.s., Slovakia, Dynatrace Master Partner

This is not working for me. I created several custom devices, one for each of the unmonitored endpoints. It successfully created a custom device and it created a separate service. The problem is that all of my custom device calls are now grouped together under the same service. I think it's because the service is tied to the process group, and a custom device uses a process group of '-'. Since they are the same, it groups them together. It looks like the workaround is not going to work. We need to be able to do something like what we can do with calls to public networks.

Agreed with Shaun here. I am seeing the same behavior after attempting this is a non-prod environment. I have heard several times from my client that seeing "Requests to unmonitored hosts" anywhere in the environment is not appealing. Especially in things like alerts or on service flow diagrams. This is especially apparent when there are multiple domains under the unmonitored host bucket. You cannot see which domain or endpoint is being called on a service flow diagram. As Shaun said, separating these out like what can be done for public networks would be a step in the right direction.

The documentation on that page is not very robust, either, and leaves a lot of the parameters purpose up for question or trial and error.

And I agree as well. I've tested this in a small environment, and after some trial and error, finding environment IDs, etc. - it works. But, it just seems like a hastle that should be easily solved via the WebUI - the same way we do it for Requests to Public Networks.

agreed, this is on the roadmap, please open an RFE.

The Service should be created per Custom Device and for multiple devices actually. the PG does not factory into it. If this is the case please report a support ticket.

I am trying to learn how to use DT API to register new custom device. I have gone through doc but not sure how to construct endpoint domain like this GET/PUT "https://{id}".

We have DT SAAS installed on internal network which we access using URL like this . I have tried appending app id but to run REST API but it did not work. I would appreciate if you can provide any guidance around it.

Dynatrace Champion
Dynatrace Champion

Featured Posts