ML
    • Recent
    • Categories
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login

    Kubernetes help. Kube-dns crashloopbackoff

    IT Discussion
    kubernetes
    4
    9
    2.0k
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • R
      ranganath-t
      last edited by scottalanmiller

      Need help with kubernetes.

      I have installed kuberentes in my centos 7.2 machine. When i type kubectl get pods --all-namespaces

      I see the kube-dns has an error saying crashloopbackoff. I am unable to find the exact problem.

      Somebody help on this ?

      1 Reply Last reply Reply Quote 1
      • JaredBuschJ
        JaredBusch
        last edited by JaredBusch

        New to CentOS 7?
        Disable SELinux and try again.

        setenforce 0

        If things work, then you can work on resolving the SELinux security concerns.

        1 Reply Last reply Reply Quote 1
        • R
          ranganath-t
          last edited by

          swapoff -a

          setenforce 0

          installed docker, kubeadm, kubelet, kubectl

          enabled and started docker

          Checked the cgroup driver for both docker and kube to be same with cgroupfs

          kubeadm init with cidr range 10.244.0.0/2

          and kubectl get pods --all-namespaces

          Kube-dns crashbackloopoff.

          JaredBuschJ 1 Reply Last reply Reply Quote 0
          • JaredBuschJ
            JaredBusch @ranganath-t
            last edited by

            @ranganath-t said in Kubernetes help. Kube-dns crashloopbackoff:

            swapoff -a

            setenforce 0

            installed docker, kubeadm, kubelet, kubectl

            enabled and started docker

            Checked the cgroup driver for both docker and kube to be same with cgroupfs

            kubeadm init with cidr range 10.244.0.0/2

            and kubectl get pods --all-namespaces

            Kube-dns crashbackloopoff.

            Did you reboot someplace in there? Because setenforce 0 does not survive a reboot.

            1 Reply Last reply Reply Quote 0
            • R
              ranganath-t
              last edited by

              I have disabled selinux=disabled.

              1 Reply Last reply Reply Quote 0
              • R
                ranganath-t
                last edited by

                fyi, i did not reboot the machine at all

                1 Reply Last reply Reply Quote 0
                • JaredBuschJ
                  JaredBusch
                  last edited by

                  Well, then I am no further help as I have not tried to use this container before.

                  1 Reply Last reply Reply Quote 0
                  • travisdh1T
                    travisdh1
                    last edited by

                    I've not used kubernetes either, but, do you have an error log from when you tried to start it? Probably in /var/log and something like kube.log.

                    Also, are any nameservers listed in /etc/resolve.conf? Network settings like to drop the DNS server setting once in a while, and the DNS error might be caused by this.

                    1 Reply Last reply Reply Quote 0
                    • stacksofplatesS
                      stacksofplates
                      last edited by

                      Did you follow this from the kubeadm install setups?

                      I've done a cluster setup a couple of times and haven't run into this.

                      1 Reply Last reply Reply Quote 0
                      • 1 / 1
                      • First post
                        Last post