15 May 2023 10:59 AM - last edited on 13 Jun 2023 04:46 PM by Ana_Kuzmenchuk
Dear Community,
I have a question regarding OpenShift and AG usage:
As per the below screenshot, my OpenShift servers are using the internal AG on Kubernetes, what is the conclusion that we can find here:
OneAgent is communicating to Kubernetes AG, and that AG forward to Dynatrace env AG, that's why we can find both on the properties of each agent?
Solved! Go to Solution.
15 May 2023 12:50 PM
Hi @Malaik,
As far as I know the contenarized AG communicate directly with your managed host without ENV AG.
Best regards,
Mizső
15 May 2023 12:58 PM
Your situation is due to (some) code modules in applications running on those hosts communicating using the non-containerized active gate.
15 May 2023 02:59 PM
Downloading the support archive is probably the easiest method to verify the assumption and verify the logs of code modules. This also depends on DynaKube deployment options you have.
You can also reach out to DTONE who have access to DebugUI and help to identify the processes and reasons why the code modules are not connected to containerized ActiveGate.
15 May 2023 06:21 PM
Did you set up any networkzone when configuring the dynakube? I do recommend to.
15 May 2023 06:22 PM
hmm, just now I saw your screenshot and NZ is there... this is odd.
31 Oct 2023 02:29 PM
I don't know if is the same problem. If in your environment you have a proxy defined to comunicate to the external dynatrace, in the kubeconfig yo have not to define a general proxy to allow the nodeagents to communicate to the internal activegate. To allow you internal activeGate to comunicate to the external dynatrace activeGates trough the proxy you have to add in kubeconfig AG definition section
customProperties: value: | [http.client] proxy-server=<your proxy> proxy-port=8080 proxy-non-proxy-hosts=*.xxx.xxx
I think is a workarround until the internal ActiveGate get the proper configuration from the dynatrace configuration to send the metrics to a internal AG