Jx: Failed to connect to kubernetes: invalid configuration: no configuration has been provided

[wangm23456]

Summary

Failed to connect to kubernetes: invalid configuration: no configuration has been provided

Jx version

The output of jx version is:

ubuntu@master:~$ jx version
Failed to connect to kubernetes: invalid configuration: no configuration has been provided
NAME        VERSION
jx          1.3.123
kubectl     v1.10.3
helm client v2.9.1+g20adb27
helm server v2.9.1+g20adb27
git         git version 2.7.4

Kubernetes cluster

What kind of Kubernetes cluster are you using & how did you create it?

run api-server etc. by systemd
export KUBECONFIG=$KUBECONFIG:$HOME/.kube/config

ubuntu@master:~$ cat .kube/config 
apiVersion: v1
kind: Config

clusters:
- cluster:
    certificate-authority: /etc/kubernetes/ssl/k8s-root-ca.csr
    server: 10.1.129.8:6443
  name: kubernetes
contexts:
- context:
    cluster: default
    user: kubectl
  name: default
- context:
    cluster: default
    namespace: jx
    user: jx
  name: jx
current-context: default
preferences: {}
users: 
- name: kubectl
  user:
    client-certificate: /etc/kubernetes/ssl/kubelet-client.crt
    client-key: /etc/kubernetes/ssl/kubelet-client.key

Operating system / Environment

ubuntu 16.04

Expected behavior

Actual behavior

CLI:

ubuntu@master:~$ jx version
Failed to connect to kubernetes: invalid configuration: no configuration has been provided
NAME        VERSION
jx          1.3.123
kubectl     v1.10.3
helm client v2.9.1+g20adb27
helm server v2.9.1+g20adb27
git         git version 2.7.4
jubuntu@master:~$ jx compliance run
error: could not create the compliance client: compliance client failed to load the Kubernetes configuration: invalid configuration: no configuration has been provided
ubuntu@master:~$ kubectl cluster-info 
Kubernetes master is running at http://localhost:8080
Elasticsearch is running at http://localhost:8080/api/v1/namespaces/kube-system/services/elasticsearch-logging/proxy
Heapster is running at http://localhost:8080/api/v1/namespaces/kube-system/services/heapster/proxy
Kibana is running at http://localhost:8080/api/v1/namespaces/kube-system/services/kibana-logging/proxy
CoreDNS is running at http://localhost:8080/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy
monitoring-grafana is running at http://localhost:8080/api/v1/namespaces/kube-system/services/monitoring-grafana/proxy
monitoring-influxdb is running at http://localhost:8080/api/v1/namespaces/kube-system/services/monitoring-influxdb/proxy

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
ubuntu@master:~$ kubectl config view 
apiVersion: v1
clusters:
- cluster:
    certificate-authority: /etc/kubernetes/ssl/k8s-root-ca.csr
    server: 10.1.129.8:6443
  name: kubernetes
contexts:
- context:
    cluster: default
    user: kubectl
  name: default
- context:
    cluster: default
    namespace: jx
    user: jx
  name: jx
current-context: default
kind: Config
preferences: {}
users:
- name: kubectl
  user:
    client-certificate: /etc/kubernetes/ssl/kubelet-client.crt
    client-key: /etc/kubernetes/ssl/kubelet-client.key

[enkicoma]

same for me

jx get git
WARNING: The current user cannot query secrets in the namespace default: Failed to create a kubernetes client invalid configuration: no configuration has been provided

[jstrachan]

what is the output of these commands:

echo $KUBECONFIG
kubectl get ns

[enkicoma]

@jstrachan
echo $KUBECONFIG
nothing.. happen
kubectl get ns
the server doesn't have a resource type "ns"

[jstrachan]

@enkicoma it sounds like you don't have a valid kubernetes cluster setup to me

[enkicoma]

@jstrachan yep, I don't because of... I want to create a git user and token for bitbucket.. jx create cluster gke --skip-login --default-admin-password=mySecretPassWord123 -n myclustername will require git username and token. If I don't have configured before this command jx will not prompt to create a new token and I will not have any repositories created at the end.

[jstrachan]

you should be able to do the jx create git server ... and jx create git token ... CLIs and ignore the errors ;) - they should still update the local ~/.jx/gitAuth.yaml file - so you can use the git server/token when creating a cluster.

[jstrachan]

ideally we'd have a CLI option like --no-cluster or something for the jx create git server|user that doesn't try to talk to the cluster to avoid the errors

[enkicoma]

@jstrachan ok, will try

jx create git user -n BitBucket someUserName -p somepassword -e [email protected]

WARNING: The current user cannot query secrets in the namespace default: Failed to create a kubernetes client invalid configuration: no configuration has been provided
error: invalid configuration: no configuration has been provided

[jstrachan]

@enkicoma did it update your ~/.jx/gitAuth.yaml file? If so you can quietly ignore that warning + error for now ;)

Am hoping we can make this a little less confusing - I think the --no-cluster flag should help lots

[enkicoma]

@jstrachan I will try today after work.. many thanks for the support!

[nagarajui7]

@jstrachan
In my case
echo $KUBECONFIG
nothing.. happen
kubectl get ns
NAME STATUS AGE
default Active 45d
kube-public Active 45d
kube-system Active 45d
pls check on this

[jstrachan]

@nagarajui7 that looks like that kubernetes cluster does not have Jenkins X installed at all? There's no jx namespace.

You probably need to install Jenkins X first: https://jenkins-x.io/getting-started/install-on-cluster/

[zhannk]

inconsistent context/cluster generated?

clusters:

• cluster:
  certificate-authority: /etc/kubernetes/ssl/k8s-root-ca.csr
  server: 10.1.129.8:6443
  name: kubernetes
  contexts:
• context:
  cluster: default
  user: kubectl
  name: default
• context:
  cluster: default
  namespace: jx
  user: jx
  name: jx
  current-context: default

[ccojocar]

It seems that the cluster context is not properly configured.

[aps554]

echo $KUBECONFIG
/etc/kubernetes/admin.conf
kubectl get ns
W0426 09:51:32.538172 1775 loader.go:223] Config not found: /etc/kubernetes/admin.conf
error: no configuration has been provided, try setting KUBERNETES_MASTER environment variable

[x10mrrobot01x]

i to am having this issue
cmd = echo $KUBECONFIG
results = :/root/.kube/config
cmd = kubectl get ns
results = W0504 02:12:47.037118 187212 loader.go:223] Config not found: /root/.kube/config
error: no configuration has been provided, try setting KUBERNETES_MASTER environment variable
much help is appreciated
thanks in advance

[ruanbekker]

I had a similar issue, where my KUBECONFIG was infact pointing to my contexts, but the mapping from context to cluster was wrong.

Where I had my env-a's context pointing to cluster: default instead of cluster: cluster-a, as mine looked like this:

clusters:
- cluster:
    certificate-authority-data: xx
    server: https://x.x.x.x:6443
  name: cluster-a
- cluster:
    certificate-authority-data: y
    server: https://y.y.y.y:6443
  name: cluster-b
contexts:
- context:
    cluster: default
    namespace: default
    user: user-a
  name: env-a
- context:
    cluster: cluster-b
    namespace: default
    user: user-b
  name: env-b

After changing this bit, it worked:

contexts:
- context:
    cluster: cluster-a
    namespace: default
    user: user-a
  name: env-a