对由kubeadm安装的kubernetes集群进行升级

最新推荐文章于 2024-05-16 19:34:29 发布

DevOps008

最新推荐文章于 2024-05-16 19:34:29 发布

阅读量1.3k

点赞数 1

分类专栏： Kubernetes 文章标签： Kubernetes

本文链接：https://blog.csdn.net/DevOps008/article/details/95591678

版权

Kubernetes 专栏收录该内容

3 篇文章 0 订阅

订阅专栏

对由kubeadm安装的kubernetes集群进行升级

kubernetes版本升级应该是一件常见的事情，kubernetes在2015年7月21号发布1.0版本，到今年2019年已经到了1.15版本，而且版本的发布速度越来越快，对于DevOps人员来讲，kubernetes版本升级也就成了一个必备技能。本文会讲述将版本从 v1.10.x 升级到 v1.11.x。升级主要包含二部分：kubeadm的升级，master节点和node节点上kubernetes 包(kubelet)的升级。

注意事项：

不能跨版本升级，只能从一个minor版本升级到下一个minor版本，比如从 v1.10.x 到 v1.11.x，不能从 v1.9.x 直接到 v1.11.x。
在 v1.11 以后默认的DNS是 CoreDNS 而不是 kube-dns，如果升级以后想继续用 kube-dns，则应该是升级的时候添加参数 kubeadm upgrade apply v1.11.x --feature-gates=CoreDNS=false

kubeadm 升级

在master节点上，按照如下命令可以完成kubeadm的升级

$ export VERSION=$(curl -sSL https://dl.k8s.io/release/stable.txt)
$ export ARCH=amd64
$ curl -sSL https://dl.k8s.io/release/${VERSION}/bin/linux/${ARCH}/kubeadm > /usr/bin/kubeadm
$ chmod a+rx /usr/bin/kubeadm

这个地方要特别注意，curl -sSL https://dl.k8s.io/release/stable.txt 会获取到kubernetes当前最稳定版本信息，目前为止是 v1.15.0，也就是意味着 kubeadm 的版本是 v1.15.0。如果按照上述方案，升级之后的 kubeadm 是 v1.15.0 ，然后在master节点上面执行 kubeadm upgrade plan 会出现如下错误

$ kubeadm upgrade plan
[upgrade/config] Making sure the configuration is correct:
[upgrade/config] Reading configuration from the cluster...
[upgrade/config] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -oyaml'
[upgrade/config] FATAL: unexpected error when reading kubeadm-config ConfigMap: ClusterConfiguration key value pair missing

原因是因为 kubeadm 的版本过高，可以通过降低版本来实现。(https://serverfault.com/questions/943696/fatal-unexpected-error-when-reading-kubeadm-config-configmap-clusterconfigurat)
。我现在的版本是 v1.10.x 我要升级到 v1.11.x, 在安装 kubeadm 的时候直接指定版本 v1.11.0 就可以解决这个问题

$ curl -sSL https://dl.k8s.io/release/v1.11.0/bin/linux/${ARCH}/kubeadm > /usr/bin/kubeadm
$ chmod a+rx /usr/bin/kubeadm

接着执行 kubeadm upgrade plan 命令

$ kubeadm upgrade plan
[preflight] Running pre-flight checks.
[upgrade] Making sure the cluster is healthy:
[upgrade/config] Making sure the configuration is correct:
[upgrade/config] Reading configuration from the cluster...
[upgrade/config] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -oyaml'
I0710 03:35:06.733248   16862 feature_gate.go:230] feature gates: &{map[]}
[upgrade] Fetching available versions to upgrade to
[upgrade/versions] Cluster version: v1.10.11
[upgrade/versions] kubeadm version: v1.11.0
[upgrade/versions] Latest stable version: v1.15.0
[upgrade/versions] Latest version in the v1.10 series: v1.10.13
[upgrade/versions] WARNING: No recommended etcd for requested kubernetes version (v1.15.0)
Components that must be upgraded manually after you have upgraded the control plane with 'kubeadm upgrade apply':
COMPONENT   CURRENT       AVAILABLE
Kubelet     4 x v1.10.4   v1.10.13
Upgrade to the latest version in the v1.10 series:
COMPONENT            CURRENT    AVAILABLE
API Server           v1.10.11   v1.10.13
Controller Manager   v1.10.11   v1.10.13
Scheduler            v1.10.11   v1.10.13
Kube Proxy           v1.10.11   v1.10.13
CoreDNS                         1.1.3
Kube DNS             1.14.8
Etcd                 3.1.12     3.1.12
You can now apply the upgrade by executing the following command:
	kubeadm upgrade apply v1.10.13
____________________________________________________________________
Components that must be upgraded manually after you have upgraded the control plane with 'kubeadm upgrade apply':
COMPONENT   CURRENT       AVAILABLE
Kubelet     4 x v1.10.4   v1.15.0
Upgrade to the latest stable version:
COMPONENT            CURRENT    AVAILABLE
API Server           v1.10.11   v1.15.0
Controller Manager   v1.10.11   v1.15.0
Scheduler            v1.10.11   v1.15.0
Kube Proxy           v1.10.11   v1.15.0
CoreDNS                         1.1.3
Kube DNS             1.14.8
Etcd                 3.1.12     N/A
You can now apply the upgrade by executing the following command:
	kubeadm upgrade apply v1.15.0
Note: Before you can perform this upgrade, you have to update kubeadm to v1.15.0.
_____________________________________________________________________

接下来执行升级命令 kubeadm upgrade apply v1.11.0。如果指定的版本高于 v1.11.x，比如 v1.12.x就会如下错

[upgrade/version] FATAL: The --version argument is invalid due to these fatal errors:
	- Specified version to upgrade to "v1.12.0" is too high; kubeadm can upgrade only 1 minor version at a time
	- Specified version to upgrade to "v1.12.0" is at least one minor release higher than the kubeadm minor release (12 > 11). Such an upgrade is not supported
Please fix the misalignments highlighted above and try upgrading again

所以不可以跨minor版本升级，只能逐级升级。

升级master节点和node节点的kubernetes包(kubelet & kubeadm)

我们先来升级master节点，先用命令(kubectl drain $HOST --ignore-daemonsets)将master节点设置为维护状态

$ kubectl drain k8stest01 --ignore-daemonsets --delete-local-data --force
node "k8stest01" cordoned
WARNING: Ignoring DaemonSet-managed pods: kube-flannel-ds-sxm2t, kube-proxy-bmg55, kube-prometheus-exporter-node-nj7nf
pod "coredns-78fcdf6894-ctvbl" evicted
node "k8stest01" drained

此时，如果可用 kubectl get nodes 查看一下master节点的状态

$ kubectl get nodes
NAME        STATUS                     ROLES     AGE       VERSION
k8stest01   Ready,SchedulingDisabled   master    203d      v1.10.4
k8stest02   Ready                      <none>    203d      v1.10.0
k8stest03   Ready                      <none>    203d      v1.10.0
k8stest04   Ready                      <none>    203d      v1.10.0

然后根据服务器的类型，选择相应的升级命令，我用的服务器是 ubuntu,所以执行下面命令

$apt-get update
$apt-get upgrade -y kubelet kubeadm

上述命令安装的 kubelet & kubeadm 都是最新版本，为了集群相匹配，可以用如下命令安装指定的版本

$ export ARCH=amd64 
$ curl -sSL https://dl.k8s.io/release/v1.11.0/bin/linux/${ARCH}/kubelet > /usr/bin/kubelet
$ chmod a+rx /usr/bin/kubelet
$ curl -sSL https://dl.k8s.io/release/v1.11.0/bin/linux/${ARCH}/kubeadm > /usr/bin/kubeadm
$ chmod a+rx /usr/bin/kubeadm

然后查看kubelet和kubeadm的版本

$ kubelet --version
Kubernetes v1.11.0
$ kubeadm version
kubeadm version: &version.Info{Major:"1", Minor:"11", GitVersion:"v1.11.0", GitCommit:"91e7b4fd31fcd3d5f436da26c980becec37ceefe", GitTreeState:"clean", BuildDate:"2018-06-27T20:14:41Z", GoVersion:"go1.10.2", Compiler:"gc", Platform:"linux/amd64"}

接着重启 kubelet 服务

$ systemctl restart kubelet
$ systemctl status kubelet
● 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 Wed 2019-07-10 09:18:33 EDT; 2h 8min ago
     Docs: https://kubernetes.io/docs/home/
 Main PID: 29813 (kubelet)
    Tasks: 21
   Memory: 54.0M
      CPU: 4min 8.792s
   CGroup: /system.slice/kubelet.service
           └─29813 /usr/bin/kubelet --bootstrap-kubeconfig=/etc/kubernetes/bootstrap-kubelet.conf --kubeconfig=/etc

接着使用命令 kubectl uncordon k8stest01 将master节点k8stest01从维护状态调整为可调度状态，然后查看版本信息我们看到master节点的版本已经从 v1.10.x 升至 v1.11.x

$ kubectl get nodes
NAME        STATUS                        ROLES    AGE    VERSION
k8stest01   Ready                         master   204d   v1.11.0
k8stest02   Ready                         <none>   204d   v1.10.0
k8stest03   Ready                         <none>   204d   v1.10.0
k8stest04   Ready                         <none>   204d   v1.10.0

至此，master节点上kubernetes的包(kubelet,kubeadm)已经升级完成。接下来升级node节点上的kubelet和kubeadm。node节点的升级和master的步骤一样，第一步也是要将node节点设置为维护状态 kubectl drain k8stest02 --ignore-daemonsets --delete-local-data --force 。接着升级kubelet和kubeadm，node节点比master多一个步骤，先执行如下命令

$ apt-get update 
$ export ARCH=amd64 
$ curl -sSL https://dl.k8s.io/release/v1.11.0/bin/linux/${ARCH}/kubelet > /usr/bin/kubelet
$ chmod a+rx /usr/bin/kubelet
$ curl -sSL https://dl.k8s.io/release/v1.11.0/bin/linux/${ARCH}/kubeadm > /usr/bin/kubeadm
$ chmod a+rx /usr/bin/kubeadm

然后执行 kubeadm upgrade node config --kubelet-version $(kubelet --version | cut -d ' ' -f 2)

$ kubeadm upgrade node config --kubelet-version $(kubelet --version | cut -d ' ' -f 2)
[kubelet] Downloading configuration for the kubelet from the "kubelet-config-1.11" ConfigMap in the kube-system namespace
[kubelet] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml"
[upgrade] The configuration for this node was successfully updated!
[upgrade] Now you should go ahead and upgrade the kubelet package using your package manager.

接着重启 kubelet 服务

$ systemctl restart kubelet
$ systemctl status kubelet
● 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 Wed 2019-07-10 09:18:33 EDT; 2h 8min ago
     Docs: https://kubernetes.io/docs/home/
 Main PID: 29813 (kubelet)
    Tasks: 21
   Memory: 54.0M
      CPU: 4min 8.792s
   CGroup: /system.slice/kubelet.service
           └─29813 /usr/bin/kubelet --bootstrap-kubeconfig=/etc/kubernetes/bootstrap-kubelet.conf --kubeconfig=/etc

接着使用命令 kubectl uncordon k8stest02 将node节点k8stest02从维护状态调整为可调度状态，然后查看版本信息

$ kubectl uncordon k8stest02
node/k8stest02 uncordoned
$ kubectl get nodes
NAME        STATUS   ROLES    AGE    VERSION
k8stest01   Ready    master   204d   v1.11.0
k8stest02   Ready    <none>   204d   v1.11.0
k8stest03   Ready    <none>   204d   v1.10.0
k8stest04   Ready    <none>   204d   v1.10.0

我们看到node节点k8stest02的版本已经从 v1.10.x 升至 v1.11.x。其他节点都可以按照上述方法进行升级。升级完成的结果如下

$ kubectl get nodes
NAME        STATUS   ROLES    AGE    VERSION
k8stest01   Ready    master   204d   v1.11.0
k8stest02   Ready    <none>   204d   v1.11.0
k8stest03   Ready    <none>   204d   v1.11.0
k8stest04   Ready    <none>   204d   v1.11.0

DevOps008

关注

1
点赞
踩
0

收藏

觉得还不错? 一键收藏
0
评论
对由kubeadm安装的kubernetes集群进行升级

对由kubeadm安装的kubernetes集群进行升级kubernetes版本升级应该是一件常见的事情，kubernetes在2015年7月21号发布1.0版本，到今年2019年已经到了1.15版本，而且版本的发布速度越来越快，对于DevOps人员来讲，kubernetes版本升级也就成了一个必备技能。本文会讲述将版本从 v1.10.x 升级到 v1.11.x。升级主要包含二部分：kubead...
复制链接

扫一扫