Question

我目前正在尝试通过创建一个Vagrant / Ansible项目来学习Kubernetes，这个项目允许我调出一个主节点和节点，并且我遇到了一些问题。我目前正在使用kubeadm来运行init和join功能。对于init，我正在解析命令行输出以获取令牌（我知道这很糟糕;是否有更标准的方法来执行此操作？）：

---
  - command: kubeadm init
    args:
      creates: /etc/kubernetes/kubelet.conf
    register: kube_init

  - local_action: copy content={{ kube_init.stdout_lines[-1].split()[-2] }} dest=../kube.token
    when: kube_init is defined

  - local_action: copy content={{ kube_init.stdout_lines[-1].split()[-1] }} dest=../kube.master_ip
    when: kube_init is defined

来源：https://gitlab.com/djsumdog/skube/blob/0e25fca668e1067a8aa4af59b019411303838e86/ansible/roles/master/tasks/main.yml

但它有效。麻烦的是，我无法连接节点：

TASK: [node | command kubeadm join --token=1a3131.637de46bfdebadd4 10.0.2.15] *** 
failed: [node1] => {"changed": true, "cmd": ["kubeadm", "join", "--token=1a3131.637de46bfdebadd4", "10.0.2.15"], "delta": "0:00:00.079545", "end": "2016-10-25 01:54:51.858545", "rc": 1, "start": "2016-10-25 01:54:51.779000", "warnings": []}
stderr: error: <node/discovery> failed to request cluster info [Get http://10.0.2.15:9898/cluster-info/v1/?token-id=1a3131: dial tcp 10.0.2.15:9898: getsockopt: connection refused]
stdout: <util/tokens> validating provided token
<node/discovery> created cluster info discovery client, requesting info from "http://10.0.2.15:9898/cluster-info/v1/?token-id=1a3131"

FATAL: all hosts have already failed -- aborting

我可以从节点VM ping我的主VM，我甚至可以ssh到它。似乎kubernetes服务也在主服务器上运行：

$ systemctl status kubelet.service 
● kubelet.service - kubelet: The Kubernetes Node Agent
   Loaded: loaded (/lib/systemd/system/kubelet.service; enabled; vendor preset: enabled)
  Drop-In: /etc/systemd/system/kubelet.service.d
           └─10-kubeadm.conf
   Active: active (running) since Tue 2016-10-25 01:51:35 UTC; 13h ago
     Docs: http://kubernetes.io/docs/
 Main PID: 5726 (kubelet)
    Tasks: 15
   Memory: 45.5M
      CPU: 11min 37.179s
   CGroup: /system.slice/kubelet.service
           ├─5726 /usr/bin/kubelet --kubeconfig=/etc/kubernetes/kubelet.conf --require-kubeconfig=true --pod-manifest-path=/etc/kubernetes/manifests --allow-privileged=true --network-plugin=cni --cni-conf-dir=/etc/cni/net.d --cni-bin-dir=/opt/cni/bin --cluster-
           └─5740 journalctl -k -f

Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532171    5726 factory.go:108] Factory "systemd" can handle container "/system.slice/var-lib-kubelet-pods-9d0e93cd\\x2d9a55\\x2d11e6\\x2db2dd\\x2d02558d159192-volumes-kubernetes.io\\x7esecret-default\\
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532175    5726 manager.go:843] ignoring container "/system.slice/var-lib-kubelet-pods-9d0e93cd\\x2d9a55\\x2d11e6\\x2db2dd\\x2d02558d159192-volumes-kubernetes.io\\x7esecret-default\\x2dtoken\\x2ddn3r9.m
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532184    5726 factory.go:104] Error trying to work out if we can handle /system.slice/run-docker-netns-default.mount: invalid container name
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532187    5726 factory.go:115] Factory "docker" was unable to handle container "/system.slice/run-docker-netns-default.mount"
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532190    5726 factory.go:108] Factory "systemd" can handle container "/system.slice/run-docker-netns-default.mount", but ignoring.
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532194    5726 manager.go:843] ignoring container "/system.slice/run-docker-netns-default.mount"
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532316    5726 factory.go:104] Error trying to work out if we can handle /system.slice/var-lib-docker-overlay-6d625f0d241bc6ae9eb243b430a71f078912a76147fc51e41e8e6e5f3226c6e9-merged.mount: error inspec
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532321    5726 factory.go:115] Factory "docker" was unable to handle container "/system.slice/var-lib-docker-overlay-6d625f0d241bc6ae9eb243b430a71f078912a76147fc51e41e8e6e5f3226c6e9-merged.mount"
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532325    5726 factory.go:108] Factory "systemd" can handle container "/system.slice/var-lib-docker-overlay-6d625f0d241bc6ae9eb243b430a71f078912a76147fc51e41e8e6e5f3226c6e9-merged.mount", but ignoring.
Oct 25 15:43:35 ubuntu-xenial kubelet[5726]: I1025 15:43:35.532330    5726 manager.go:843] ignoring container "/system.slice/var-lib-docker-overlay-6d625f0d241bc6ae9eb243b430a71f078912a76147fc51e41e8e6e5f3226c6e9-merged.mount"

为什么我的主人不听连接？这是我正在使用的确切提交：

https://gitlab.com/djsumdog/skube/tree/0e25fca668e1067a8aa4af59b019411303838e86

如果你安装了ansible和vagrant，你应该可以运行vagrant up并获得相同的结果。

Answer 1

事实证明这是由于我的Vagrant设置。两个虚拟机都获得了相同的IP地址。我需要在我的Vagrantfile中指定:private_network和ip:：

...
config.vm.define "master" do |master|
  master.vm.network :private_network, ip: 10.10.30.10
...
...
config.vm.define "node1" do |node1|
  node1.vm.network :private_network, :ip: 10.10.30.50
...

Answer 2

尝试在 kubeadm init 之后和 kubeadm join 之前执行以下命令

  <select >
    <?php
    $args = array( 'post_type'=>'post');
    $myposts = get_posts( $args );
    foreach ( $myposts as $post ) : setup_postdata( $post ); ?>
        <option value="<?php the_ID();?>"><?php the_title(); ?></option>
    <?php endforeach; 
    wp_reset_postdata();?>
    </select>

请参阅此页http://kubernetes.io/docs/getting-started-guides/kubeadm/

无法在Vagrant / Ansible中获取kubernetes节点以连接到master

2 个答案: