lab 13.3 - charts on dashboard

mariuszr

Hello,

in guide for lab 13.3 (dashboard configuration) - on figure 13.3 there are beautiful charts for CPU and memory usage. On my dashboard I've everything except those charts. What I'm missing in configuration?

Regards,
Mariusz

chrispokorni

Hi @mariuszr,

Assuming that your metrics-server is operational, from any of the following Dashboard view you should see the CPU Usage and Memory Usage charts: Overview, Deployments, Pods, etc... These views can be selected from the left navigation panel.

Regards,
-Chris

mariuszr

Hi Chris,

Yes, metrics-server is working fine.

here how it looks in my case

Regards,
Mariusz

chrispokorni

Hi @mariuszr,

Did you happen to install the dashboard from a helm chart? Can you confirm that the dashaccess clusterrolebinding binds the correct serviceaccount with the correct clusterrole?

Also, is the metrics-server deployment configured with the additional --kubelet-... arguments as presented in step 5?

Regards,
-Chris

mariuszr

Hi Chris,

yes, dashboard installed from helm chart (with service modified to NodePort)
yes, roles are correct:
Name: dashaccess
Labels:
Annotations:
Role:
Kind: ClusterRole
Name: cluster-admin
Subjects:
Kind Name Namespace
---- ---- ---------
ServiceAccount dashboard-kubernetes-dashboard default___

and, one more yes, metrics-server deployment had both --kubelet-... arguments - bellow piece of describe command
"Containers:
metrics-server:
Image: k8s.gcr.io/metrics-server/metrics-server:v0.3.7
Port: 4443/TCP
Host Port: 0/TCP
Args:
--cert-dir=/tmp
--secure-port=4443
--kubelet-insecure-tls
--kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname"

Regards,
Mariusz

chrispokorni

Hi @mariuszr,

I have seen similar issues in the past with misconfigured Dashboard charts. In those situations the recommendation was to install the Dashboard following the documentation instead. As a result, steps 1 through 4 of the "Configure the Dashboard" lab should be replaced by:

The installation command from the documentation:
kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.2.0/aio/deploy/recommended.yaml

Edit the service and convert it into a NodePort type service:
kubectl -n kubernetes-dashboard edit svc kubernetes-dashboard

List the service to view the value of the node port (3xxxx):
kubectl -n kubernetes-dashboard get svc kubernetes-dashboard

Give the Dashboard admin access (edit and update the existing clusterrolebinding, or replace it with a new one):
kubectl create clusterrolebinding dashaccess --clusterrole=cluster-admin --serviceaccount=kubernetes-dashboard:kubernetes-dashboard

Regards,
-Chris

maybel

Hi @chrispokorni, I followed all the steps above; it looks like I did right, except for the secret results, and the service account dashboard-kubernetes-dashboard is not in the default namespace. I guess I'm missing something.

$ kubectl -n kubernetes-dashboard get svc kubernetes-dashboard
NAME                   TYPE       CLUSTER-IP      EXTERNAL-IP   PORT(S)         AGE
kubernetes-dashboard   NodePort   10.99.194.242   <none>        443:30951/TCP   5m47s
[email protected]:~$ kubectl create clusterrolebinding dashaccess --clusterrole=cluster-admin --serviceaccount=kubernetes-dahboard:kubernetes-dashboard
clusterrolebinding.rbac.authorization.k8s.io/dashaccess created
[email protected]:~$ kubectl get serviceaccounts
NAME                      SECRETS   AGE
default                   0         64d
myingress-ingress-nginx   0         42d
[email protected]:~$ kubectl -n kubernetes-dashboard get serviceaccounts
NAME                   SECRETS   AGE
default                0         28m
kubernetes-dashboard   0         28m

maybel

When I tried to open the Kubernetes dashboard, my browser crashed.

maybel

I created recommended.yaml file from the link and modified its type.

[email protected]:~$ kubectl apply -f recommended.yaml
namespace/kubernetes-dashboard created
serviceaccount/kubernetes-dashboard created
service/kubernetes-dashboard created
secret/kubernetes-dashboard-certs created
secret/kubernetes-dashboard-csrf created
secret/kubernetes-dashboard-key-holder created
configmap/kubernetes-dashboard-settings created
role.rbac.authorization.k8s.io/kubernetes-dashboard created
clusterrole.rbac.authorization.k8s.io/kubernetes-dashboard created
rolebinding.rbac.authorization.k8s.io/kubernetes-dashboard created
clusterrolebinding.rbac.authorization.k8s.io/kubernetes-dashboard created
deployment.apps/kubernetes-dashboard created
service/dashboard-metrics-scraper created
Warning: spec.template.metadata.annotations[seccomp.security.alpha.kubernetes.io/pod]: deprecated since v1.19, non-fuctional in a future release; use the "seccompProfile" field instead
deployment.apps/dashboard-metrics-scraper created

chrispokorni

Hi @maybel,

The alternate installation method above was posted in 2021. Since then, the dashboard helm chart has been quite stable. From your comments above, it seems you may be mixing instructions from the lab guide with the ones from my older comment. For consistency I would recommend following the lab guide to install and configure the dashboard chart.

As far as the browser crashing, I cannot reproduce that, but I know in Chrome you may need to click the advanced button to allow the browser to display unsecured content.

Regards,
-Chris

maybel

Thank you, @chrispokorni. I am working on that.

maybel

Hi @chrispokorni, I followed the lab guide. I got all the service accounts but zero secrets, and I wonder how I can fix it.

Get the Kubernetes Dashboard URL by running:
  export POD_NAME=$(kubectl get pods -n default -l "app.kubernetes.io/name=kubernetes-dashboard,app.kubernetes.io/instance=k8s-dashboard" -o jsonpath="{.items[0].metadata.name}")
  echo https://127.0.0.1:8443/
  kubectl -n default port-forward $POD_NAME 8443:8443
[email protected]:~$ kubectl get serviceaccounts
NAME                                 SECRETS   AGE
default                              0         75d
k8s-dashboard-kubernetes-dashboard   0         3m21s
myingress-ingress-nginx              0         53d
[email protected]:~$

maybel

Another problem is the lab said:

On your local system, open a browser and navigate to an HTTPS URL made of the Public IP and the high-numbered port...

And I have one port, 443.

chrispokorni

Hi @maybel,

I would not worry about the secrets counts being "0". It is expected in latest k8s releases. Feel free to proceed with the clusterrolebinding and the token creation steps.

If the values.yaml edit was successful in step 2, then there should be a NodePort type service for the dashboard release, listing 2 distinct port values: an internal port 443 and an external high-valued node port (randomly assigned between 30000 and 32767).

Regards,
-Chris