What is the exact difference between infrastructure only monitoring and full stack in Dynatrace?

[larry_roberts]

I am trying to find documentation that goes into detail on the exact differences between infrastructure only monitoring and full stack and can't seem to find any. I have the full stack down, but not exactly sure what all infrastructure only will capture.

Thanks!

[Joe_Hoffman · Accepted solution]

Here's the doc page which describes Infra-only monitoring:

https://www.dynatrace.com/support/help/monitor/hos...

Basically, infra-only monitoring captures the infrastructure metrics, such as disk, IO, CPU, memory, process list, etc. But does not capture application details or user experience data. So for example, with Infra-only you would not be able to drill down into the details of transactions that are running thru your Java,.NET, Go, etc. processes, but you would be able to see how much CPU, Memory, Disk IO was being consumed by these processes.

Let us know if this answers your question.

[larry_roberts]

Thank you!

[MAkimov]

@Joseph M. H. Hello

Is OneAgent Plugin working for Infra-only monitoring ?

[Joe_Hoffman]

Plugins can run on both infra-only OneAgents as well as full OneAgents.

Here's a doc page that states this fact.

https://www.dynatrace.com/support/help/extend-dynatrace/plugins/oneagent-plugins/how-to-create-a-pyt...

[kalle_lahtinen]

Hi,

I think this is a good question. On a Dynatrace UI settings level, one could maybe say that the definition of having full-stack enabled simply means that it allows you to turn on deep monitoring for processes. That in turn leads into various process-specific insights, real user session data, code-level details, etc. - depending on the process.

But let's say you have a server with only one vital process (worth monitoring) running. And for that, there's a compatibility issue with OneAgent, which means you have to disable deep monitoring for that particular process. In this case it would appear there is no reason to keep full-stack mode on for that server, since you won't have deep monitoring enabled for any of the processes. Is my assumption here correct?

[skrystosik]

Yes in such case you can switch to infrastructure and you will see metrics from hosts and processes without deep monitoring.

Sebastian

[larry_roberts]

I agree it's not exactly clear. While I understanding it all now, I think it still confuses allot of people who are new to Dynatrace and could be much better explained through better documentation.

[Joe_Hoffman]

Larry, I"m happy to drive forward some improved doc changes, but I'm not sure what doc changes would help clarify. I'm looking for your suggestions.

There's two key pages that describes the differences:

1) https://www.dynatrace.com/support/help/how-to-use-dynatrace/hosts/basic-concepts/how-effective-is-in...

2) https://www.dynatrace.com/support/help/how-to-use-dynatrace/hosts/basic-concepts/get-started-with-in...

Can you clarify what could be done to clarify the difference between Infra and Full mode?

[MarwanC]

This might help for the management team at your organisation. It would really help if you can pin point exactly in documentation what Dynatrace do to your application top (RUM) to down approach in order to achieve those objectives and this is exactly what is missing and Larry is seeking.

Infrastructure Monitoring:

• Focuses on monitoring the underlying hardware and software components of your IT infrastructure.
• Monitors servers, networks, storage devices, and other infrastructure elements.
• Provides insights into resource utilization, performance metrics, and health status of individual infrastructure components.
• Helps identify issues related to hardware failures, resource bottlenecks, and infrastructure scalability.

Full-Stack Monitoring:

• Offers a comprehensive view of the entire application stack, including both infrastructure and application layers.
• Monitors not only the infrastructure components but also the application code, databases, APIs, and user experience.
• Provides insights into end-to-end performance, including response times, error rates, and dependencies between different components.
• Enables tracking of application performance across distributed environments
• Helps identify issues impacting the overall user experience, such as slow page load times, transaction failures, and database bottlenecks.

Summary

While infrastructure monitoring focuses on the health and performance of underlying hardware and software, full-stack monitoring offers a more holistic view that includes application-level insights and end-user experience monitoring.