error marking master: timed out waiting for the condition [kubernetes]

Question

Just I am starting to learn Kubernetes. I\'ve installed CentOS 7.5 with SELinux disabled kubectl, kubeadm and kubelet by Kubernetes YUM repository.

However, when I w

Answer 1

You are hitting the following issue in kubernetes

https://github.com/kubernetes/kubeadm/issues/1092

The workaround is to provide --node-name=<hostname> . Just go through the above ticket for more info. Hope this helps

EDIT: I have the same issue in kubeadm-1.10.0 After removing --hostname-override from /etc/systemd/system/kubelet.service.d/10-kubeadm.conf file, atleast able to initialize cluster. Didn't give provide --node-name in my cluster

Answer 2

I would recommend to bootstrap Kubernetes cluster as guided in the official documentation. I've proceeded with some steps to build cluster on the same CentOS version CentOS Linux release 7.5.1804 (Core) and will share them with you, hope it can be helpful to you to get rid of the issue during installation.

First wipe your current cluster installation:

# kubeadm reset -f && rm -rf /etc/kubernetes/

Add Kubernetes repo for further kubeadm, kubelet, kubectl installation:

[kubernetes]
name=Kubernetes
baseurl=https://packages.cloud.google.com/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://packages.cloud.google.com/yum/doc/yum-key.gpg https://packages.cloud.google.com/yum/doc/rpm-package-key.gpg
exclude=kube*
EOF

Check whether SELinux is in permissive mode:

# getenforce
Permissive

Ensure net.bridge.bridge-nf-call-iptables is set to 1 in your sysctl:

# cat <<EOF >  /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF
sysctl --system

Install required Kubernetes components and start services:

# yum update && yum upgrade && yum install -y docker kubelet kubeadm kubectl --disableexcludes=kubernetes

# systemctl start docker kubelet && systemctl enable docker kubelet

Deploy the cluster via kubeadm:

kubeadm init --pod-network-cidr=10.244.0.0/16

I prefer to install Flannel as the main CNI in my cluster, although there are some prerequisites for proper Pod network installation, I've passed --pod-network-cidr=10.244.0.0/16 flag to kubeadm init command.

Create Kubernetes Home directory for your user and store config file:

$ mkdir -p $HOME/.kube
$ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
$ sudo chown $(id -u):$(id -g) $HOME/.kube/config

Install Pod network, in my case it was Flannel:

$ kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/bc79dd1505b0c8681ece4de4c0d86c5cd2643275/Documentation/kube-flannel.yml

Finally check Kubernetes core Pods status:

$ kubectl get pods --all-namespaces

NAMESPACE     NAME                                 READY   STATUS    RESTARTS   AGE
kube-system   coredns-576cbf47c7-4x7zq             1/1     Running   0          36m
kube-system   coredns-576cbf47c7-666jm             1/1     Running   0          36m
kube-system   etcd-centos-7-5                      1/1     Running   0          35m
kube-system   kube-apiserver-centos-7-5            1/1     Running   0          35m
kube-system   kube-controller-manager-centos-7-5   1/1     Running   0          35m
kube-system   kube-flannel-ds-amd64-2bmw9          1/1     Running   0          33m
kube-system   kube-proxy-pcgw8                     1/1     Running   0          36m
kube-system   kube-scheduler-centos-7-5            1/1     Running   0          35m

In case you still have any doubts, just write down a comment below this answer.