18 Jul 2018 02:34 AM - last edited on 23 Feb 2023 08:13 AM by Karolina_Linda
Hi there 🙂
As a colleague of mine, who was responsible for our dynatrace managed, is not available any longer the honor to operate it is now mine.
But I somehow miss a administrator documentation for dynatrace manged.
E.g. what exactly does the cassandra do, what is the metric storage, what can I see and do on the server itself.
Or, another example, these storage types are not evenly distributed. What can I do?
So all the things I don't really do via the GUI.
Is there some kind of documentation for this? This would be really helpful.
Thanks a lot 🙂
Solved! Go to Solution.
I'm not aware of any extra docs except the official Dynatrace Help. However, you can go through the Dynatrace University modules called "Dynatrace for Administrators" and take a look at the Dynatrace Managed section here.
There are also many admin experts here in the forums, so I hope your specific questions will be answered quickly 🙂
Actually, there is no public documentation on the Dynatrace server internals and I don't think that will change in the future. Main reason for that is, that is bundled as a single product and you don't have to care about the Dynatrace managed server components at all.
If something breaks - you (as admin) will get notified via email or you can see that in the Cluster Management Console.
So I'd recommend not to try to modify and tune anything in any of the internal components (unless support instructs you to do so) and just enjoy the product as it is :-).
Thanks to both of you 🙂
I thought there would be some "secret" documentation 😄
Just basics like: as an administrator you should know that metrics data is stored in elastic and something like this 🙂
Or what to do when data is not split evenly between nodes.
But then I'll just bother the support in case I run into some trouble.
Dynatrace Managed as the name of the product obliges, comes with a Mission Control team that looks after your cluster. Your goal as an administrator is:
* Watch cluster notifications, and react on them
* Provide enough disk space for each of data stores: Transaction storage, Cassandra (metrics/timeseries), Elasticsearch and optionally additional storage - backup, log analytics
* Keep infrastructure healthy
* Maintain cluster configuration and user authentication
Here's a nice overview how the architecture of Dynatrace Managed looks like:
In short, Cluster is built from nodes that are same. Each node contains Nginx, Server, Security Gateway, Cassandra and Elasticsearch.
Server and Security Gateway process OneAgent traffic and store the transaction data in Transaction Storage, Cassandra and Elasticsearch accordingly. All components should equally distribute the load among all nodes.
That should be a good start to you. Don't hesitate to ask any specific questions that comes to you. Remember, that Community and Dynatrace One help chat is here to help you any time you need.
Thanks for your answer, Radoslaw 🙂
I'll add this to my own documentation right away. It's good to know which component (e.g. elastic) is responsible for what thing. Just if something does not feel right I can point the support to the right direction.