Uploaded image for project: 'Couchbase Kubernetes'
  1. Couchbase Kubernetes
  2. K8S-2363

Manually bootstrapped on-prem clusters do not work

    XMLWordPrintable

Details

    • Bug
    • Status: Open
    • Minor
    • Resolution: Unresolved
    • None
    • None
    • None
    • Linux CentOS 7 or CentOS 8 with open source Kubernetes 1.21 or 1.22 and Docker 20.10 running on vSphere.
    • 1

    Description

      Manually bootstrapped on-prem Kubernetes clusters built with kubeadm init do not work. Server pods never pass the readiness test as port 8091 is unreachable. Several different Linux distributions were tried, along with k8s 1.21 and 1.22, with Calico or Flannel. NGINX Ingress and MetalLB where also tried to no avail. The same result was observed with or without the Ingress, with or without the load balancer. One or two Couchbase Server pods are created and never pass the readiness test an are eventually deleted and a new pod is tried. However in this same environment, on the same Linux templates, Rancher deployed clusters work fine.
       
      High level steps to reproduce the issue: # Create CentOS 7 system based on minimal software profile

      1. Configure EPEL repo
      2. Configure docker-ce repo
      3. Set docker to use systemd cgroup

      # yum install kubelet-1.21.4-0 kubeadm-1.21.4-0 kubectl-1.21.4-0 --disableexcludes=kubernetes
      # systemctl enable kubelet
      # kubeadm init --pod-network-cidr=10.64.0.0/16
      $ mkdir -p $HOME/.kube
      $ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
      $ sudo chown $(id -u):$(id -g) $HOME/.kube/config
      $ kubectl create -f }}{{https://docs.projectcalico.org/manifests/tigera-operator.yaml
      $ curl }}{{https://docs.projectcalico.org/manifests/custom-resources.yaml{{ -O}}
      $ vi custom-resources.yaml # change default CIDR to 10.64.0.0/16
      $ kubectl create -f custom-resources.yaml
      $ kubectl taint nodes --all }}{{node-role.kubernetes.io/master-
      $ kubectl get nodes # wait for ready state
      $ cd couchbase-autonomous-operator-kubernetes_2.2.1-linux-x86_64
      $ kubectl create -f crd.yaml
      $ bin/cbopcfg create admission
      $ bin/cbopcfg create operator
      $ kubectl create -f couchbase-cluster.yaml
       
      This is only one simple variation that I tried. I got the same result with multi-node manual bootstrapped clusters. And I also tried adding NGINX Ingress and MetalLB on some tests to no avail.

      Attachments

        No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

          There are no comments yet on this issue.

          People

            simon.murray Simon Murray
            michael.minichino Michael Minichino
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:

              Gerrit Reviews

                There are no open Gerrit changes

                PagerDuty