We have followed the latest improvement about monitoring Kubernetes and definitivelly better than it used to be. Thanks 🙂
I have one question about pod/container monitoring. We use PGBouncer to connect to our Postgresql SQL Database.
I can see the pod in cloud application, i can see the CPU and the ram usage. Perfectly fine but can't get any historic data like others containers/pods.
If i click on it I get that :
But I am sure you have the historic data (cpu/ram/...) as you displayed them in the first list.
How can I get those historic data ?
Note : We have one agent on all our kubernetes nodes and we are connected with activegate to the Kubernetes API (and we are up to date on both).
Exactly ! We deploy the oneagent with Dynatrace Operator and we are all up to date.
For PHP/Nginx Process I can see the metrics without any issue but not pgbouncer. But I don't think PGbouncer is supported in full stack.
But from my understand i should be able to see the CPU usage/Ram usage of this container/pod.
So I can see deployment in cloud application and if I click on one deployment, i will see the list of pod inside my container. But is it only true if the pod tehchology is supported by Dynatrace.
So if I sum up :
- the data of cloud applications is coming from kubernetes API (whatever the technology)
- If i click on one deployement in cloud application, the data displayed in "technology overview" come from OneAgent installed in AKS (so need to be supported by Dynatrace)
If i want to see the cpu usage/memory usage/throtthling, i need to go to each node, go to View Container.
=> But with this screen I don't have any historic (only the last 30 minutes !).
Is there any way to have history of cpu/memory/thotthling on the last 7 days for one pod or 1 container ? I though it was in the scope of all the improvement done about Kubernetes but it seems not.