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

kubernetes dashboard not loading

gust_henk
Newcomer_

I have started my trial and got my AWS EKS cluster connected to my Dynatrace envrionment.

 

Screenshot 2023-11-09 at 11.09.40 AM.png

Next, I tried to load one of the JSON dashobards from Andy's Github post:

https://github.com/dynatrace-perfclinics/dynatrace-getting-started/tree/main

 

Specifically I am trying to load the "Kubernetes Cluster Overview" dashboard.  When I load the JSON file, the dashboard is coming up but with all tiles erroring out:

 

Screenshot 2023-11-09 at 11.13.13 AM.png

 

Can anyone help figure out why this dashboard doesn't load for me?

8 REPLIES 8

Mizső
DynaMight Guru
DynaMight Guru

Hi @gust_henk,

I made a copy in the Playgorund about Andi dashboard in order to be able edit it.

Mizs_0-1699568719914.png

As I see Andi used a variable for the cluster name, so you should see your cluster name among the values:

Mizs_2-1699569096129.png

Click on the cluster_name under Name, so this is the query behind it, run the query, do you have a result?:

Mizs_3-1699569200101.png

Extra question: Have you connected to the local or public endpoint of the cluster?

Mizs_4-1699569503532.png

I hope it helps.

Best regards,

Mizső

Dynatrace Community RockStar 2024, Certified Dynatrace Professional

Exactly, you still need to configure the connection as @Mizső wrote. I checked with me on the environment and it works.

Have a nice day!

I also wanted to point out that when I open the query in my version of Andy's dashboard (latest clone from github), my query is different from the one show in the screen shot above:

 

timeseries sum(dt.kubernetes.nodes), by: {k8s.cluster.name}, interval: 1d
| fields k8s.cluster.name

My connection screen looks different than the image you've posted:

It looks like you are showing the "Connect containerized ActiveGate to local kubernetes API endpoint" as turned off, whereas in my Dynatrace, that button is turned on and the connection is successful:

 

Screenshot 2023-11-10 at 9.02.32 AM.png

But even as this connection shows success, I still do not have the "cluster_name" variable returning any data:

 

Screenshot 2023-11-10 at 9.06.16 AM.png

I re-imported this dashboard and got the same error as you regardless of this set connection.

Zrzut ekranu 2023-11-12 o 16.39.31.png

@andreas_grabner  can you give us a hint what we are doing wrong?

Have a nice day!

andreas_grabner
Dynatrace Guru
Dynatrace Guru

Hi. You are right - the Kubernetes Dashboard is a bit out-of-date. I created it a while ago. In the meantime some of those metrics I used have been renamed. Let me work on this and update the dashboard in both GitHub as well as on the Playgrond Tenant

One more thing on Kubernetes. We just announced the new Kubernetes Observability Experience with our new Kubernetes App. It will be GA'd in a couple of sprints. Good news is that this app will also come with some new out-of-the-box dashboards itself. So - this dashboard that I provided might no longer be needed in the future. 

But - let me work on fixing it. As I said - it should mainly be renaming of the metrics as we have renamed some of those k8s metrics in the past couple of weeks

Contact our DevRel team through devrel@dynatrace.com

@andreas_grabner  super, thank you very much for your answer.
We will be waiting 😊

Have a nice day!

Thank you Andi!  This was driving me bat-sh*@ crazy, 🙂

Featured Posts