kubeadm是Kubernetes官方提供的用于快速安部署Kubernetes集群的工具。伴随Kubernetes每个版本的发布，kubeadm可能会对集群配置方面的一些实践做调整，通过实验kubeadm可以学习到Kubernetes官方在集群配置上一些新的最佳实践。
1.准备
1.1 系统配置
在安装之前，需要先做好如下准备。
3台Linux主机如下：
- node4 - Ubuntu 22.04
- node5 - openEuler release 22.03 (LTS-SP2)
- node6 - Rocky Linux release 8.8 (Green Obsidian)
cat /etc/hosts
192.168.96.154 node4
192.168.96.155 node5
192.168.96.156 node6在各个主机上完成下面的系统配置。
如果系统启用了selinux，使用下面的命令禁用selinux:
setenforce 0
vi /etc/selinux/config
SELINUX=disabled如果各个主机启用了防火墙策略，需要开放Kubernetes各个组件所需要的端口，可以查看Ports and Protocols中的内容, 开放相关端口或者关闭主机的防火墙。
创建/etc/modules-load.d/containerd.conf配置文件，确保在系统启动时自动加载所需的内核模块，以满足容器运行时的要求:
cat << EOF > /etc/modules-load.d/containerd.conf
overlay
br_netfilter
EOF执行以下命令使配置生效:
modprobe overlay
modprobe br_netfilter创建/etc/sysctl.d/99-kubernetes-cri.conf配置文件：
cat << EOF > /etc/sysctl.d/99-kubernetes-cri.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.ipv4.ip_forward = 1
user.max_user_namespaces=28633
EOF执行以下命令使配置生效:
sysctl -p /etc/sysctl.d/99-kubernetes-cri.conf在文件名
/etc/sysctl.d/99-kubernetes-cri.conf中，“99” 代表文件的优先级或顺序。sysctl是Linux内核参数的配置工具，它可以通过修改/proc/sys/目录下的文件来设置内核参数。在/etc/sysctl.d/目录中，可以放置一系列的配置文件，以便在系统启动时自动加载这些参数。这些配置文件按照文件名的字母顺序逐个加载。数字前缀用于指定加载的顺序，较小的数字表示较高的优先级。
1.2 配置服务器支持开启ipvs的前提条件
由于ipvs已经加入到了内核的主干，所以为kube-proxy开启ipvs的前提需要加载以下的内核模块：
ip_vs
ip_vs_rr
ip_vs_wrr
ip_vs_sh
nf_conntrack_ipv4创建/etc/modules-load.d/ipvs.conf文件，保证在节点重启后能自动加载所需模块:
cat > /etc/modules-load.d/ipvs.conf <<EOF
ip_vs
ip_vs_rr
ip_vs_wrr
ip_vs_sh
EOF执行以下命令使配置立即生效:
modprobe ip_vs
modprobe ip_vs_rr
modprobe ip_vs_wrr
modprobe ip_vs_sh使用lsmod | grep -e ip_vs -e nf_conntrack命令查看是否已经正确加载所需的内核模块。
接下来还需要确保各个节点上已经安装了ipset软件包，为了便于查看ipvs的代理规则，最好安装一下管理工具ipvsadm。
在Ubuntu系统上执行:
apt install -y ipset ipvsadm在openEuler或Rocky Linux系统上执行:
yum install -y ipset ipvsadm如果不满足以上前提条件，即使kube-proxy的配置开启了ipvs模式，也会退回到iptables模式。
1.3 部署容器运行时Containerd
在各个服务器节点上安装容器运行时Containerd。
下载Containerd的二进制包， 需要注意cri-containerd-(cni-)-VERSION-OS-ARCH.tar.gz发行包自containerd 1.6版本起已经被弃用，在某些 Linux 发行版上无法正常工作，并将在containerd 2.0版本中移除，这里下载containerd-<VERSION>-<OS>-<ARCH>.tar.gz的发行包，后边再单独下载安装runc和CNI plugins:
wget https://github.com/containerd/containerd/releases/download/v1.7.11/containerd-1.7.11-linux-amd64.tar.gz将其解压缩到/usr/local下:
tar Cxzvf /usr/local containerd-1.7.11-linux-amd64.tar.gz
bin/
bin/containerd-shim-runc-v2
bin/ctr
bin/containerd-shim
bin/containerd-shim-runc-v1
bin/containerd-stress
bin/containerd接下来从runc的github上单独下载安装runc，该二进制文件是静态构建的，并且应该适用于任何Linux发行版。
wget https://github.com/opencontainers/runc/releases/download/v1.1.9/runc.amd64
install -m 755 runc.amd64 /usr/local/sbin/runc接下来生成containerd的配置文件:
mkdir -p /etc/containerd
containerd config default > /etc/containerd/config.toml根据文档Container runtimes中的内容，对于使用systemd作为init system的Linux的发行版，使用systemd作为容器的cgroup driver可以确保服务器节点在资源紧张的情况更加稳定，因此这里配置各个节点上containerd的cgroup driver为systemd。
修改前面生成的配置文件/etc/containerd/config.toml：
[plugins."io.containerd.grpc.v1.cri".containerd.runtimes.runc]
...
[plugins."io.containerd.grpc.v1.cri".containerd.runtimes.runc.options]
SystemdCgroup = true再修改/etc/containerd/config.toml中的
[plugins."io.containerd.grpc.v1.cri"]
...
# sandbox_image = "registry.k8s.io/pause:3.8"
sandbox_image = "registry.aliyuncs.com/google_containers/pause:3.9"为了通过systemd启动containerd，请还需要从https://raw.githubusercontent.com/containerd/containerd/main/containerd.service下载containerd.service单元文件，并将其放置在 /etc/systemd/system/containerd.service中。
cat << EOF > /etc/systemd/system/containerd.service
[Unit]
Description=containerd container runtime
Documentation=https://containerd.io
After=network.target local-fs.target
[Service]
ExecStartPre=-/sbin/modprobe overlay
ExecStart=/usr/local/bin/containerd
Type=notify
Delegate=yes
KillMode=process
Restart=always
RestartSec=5
# Having non-zero Limit*s causes performance problems due to accounting overhead
# in the kernel. We recommend using cgroups to do container-local accounting.
LimitNPROC=infinity
LimitCORE=infinity
# Comment TasksMax if your systemd version does not supports it.
# Only systemd 226 and above support this version.
TasksMax=infinity
OOMScoreAdjust=-999
[Install]
WantedBy=multi-user.target
EOF配置containerd开机启动，并启动containerd，执行以下命令:
systemctl daemon-reload
systemctl enable containerd --now
systemctl status containerd下载安装crictl工具:
wget https://github.com/kubernetes-sigs/cri-tools/releases/download/v1.29.0/crictl-v1.29.0-linux-amd64.tar.gz
tar -zxvf crictl-v1.29.0-linux-amd64.tar.gz
install -m 755 crictl /usr/local/bin/crictl使用crictl测试一下，确保可以打印出版本信息并且没有错误信息输出:
crictl --runtime-endpoint=unix:///run/containerd/containerd.sock version
Version: 0.1.0
RuntimeName: containerd
RuntimeVersion: v1.7.11
RuntimeApiVersion: v12.使用kubeadm部署Kubernetes
2.1 安装kubeadm和kubelet
下面在各节点安装kubeadm和kubelet：
在Ubuntu系统上执行下面的命令:
apt-get update
apt-get install -y apt-transport-https ca-certificates curl gpg
curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.29/deb/Release.key | sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg
echo 'deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] https://pkgs.k8s.io/core:/stable:/v1.29/deb/ /' | sudo tee /etc/apt/sources.list.d/kubernetes.list
apt-get update
apt install kubelet kubeadm kubectl
apt-mark hold kubelet kubeadm kubectl在openEuler和Rocky Linux系统中执行下面的命令:
cat <<EOF | sudo tee /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=https://pkgs.k8s.io/core:/stable:/v1.29/rpm/
enabled=1
gpgcheck=1
gpgkey=https://pkgs.k8s.io/core:/stable:/v1.29/rpm/repodata/repomd.xml.key
exclude=kubelet kubeadm kubectl cri-tools kubernetes-cni
EOF
yum makecache
yum install -y kubelet kubeadm kubectl --disableexcludes=kubernetes运行kubelet --help可以看到原来kubelet的绝大多数命令行flag参数都被DEPRECATED了，官方推荐我们使用--config指定配置文件，并在配置文件中指定原来这些flag所配置的内容。具体内容可以查看这里Set Kubelet parameters via a config file。最初Kubernetes这么做是为了支持动态Kubelet配置（Dynamic Kubelet Configuration），但动态Kubelet配置特性从k8s 1.22中已弃用，并在1.24中被移除。如果需要调整集群汇总所有节点kubelet的配置，还是推荐使用ansible等工具将配置分发到各个节点。
kubelet的配置文件必须是json或yaml格式，具体可查看这里。
在Kubernetes 1.22版本之前，Kubernetes没有为Linux系统提供NodeSwap功能，即默认要求关闭系统的Swap，如果不关闭，默认配置下kubelet将无法启动。
从Kubernetes 1.22开始引入了NodeSwap的Alpha支持，改功能在Kubernetes 1.28进入了Beta。
这里我们将激活NodeSwap功能，可以通过在启用NodeSwap feature gate来在节点上启用使用交换内存。此外，必须禁用failSwapOn配置设置，或者必须停用已弃用的--fail-swap-on命令行标志。
可以配置memorySwap.swapBehavior选项，以定义节点利用交换内存的方式。例如：
# 此片段放入kubelet的配置文件
memorySwap:
swapBehavior: UnlimitedSwapswapBehavior的可用配置选项包括：
UnlimitedSwap（默认）：Kubernetes工作负载可以使用它们请求的所有交换内存，最多达到系统限制。
LimitedSwap：Kubernetes工作负载对交换内存的利用受到限制。只允许使用Burstable QoS(可突发QoS)的Pod使用Swap。
如果没有为memorySwap配置，并且启用了NodeSwap特性门，kubelet将默认应用与UnlimitedSwap设置相同的行为。
注意，NodeSwap仅支持cgroup v2。对于Kubernetes v1.28，不再支持与cgroup v1一起使用swap。
可以使用下面命令查看系统支持的cgroup版本:
grep cgroup /proc/filesystems
nodev cgroup
nodev cgroup22.2 使用kubeadm init初始化集群
在各节点开机启动kubelet服务：
systemctl enable kubelet.service使用kubeadm config print init-defaults --component-configs KubeletConfiguration可以打印集群初始化默认的使用的配置：
apiVersion: kubeadm.k8s.io/v1beta3
bootstrapTokens:
- groups:
- system:bootstrappers:kubeadm:default-node-token
token: abcdef.0123456789abcdef
ttl: 24h0m0s
usages:
- signing
- authentication
kind: InitConfiguration
localAPIEndpoint:
advertiseAddress: 1.2.3.4
bindPort: 6443
nodeRegistration:
criSocket: unix:///var/run/containerd/containerd.sock
imagePullPolicy: IfNotPresent
name: node
taints: null
---
apiServer:
timeoutForControlPlane: 4m0s
apiVersion: kubeadm.k8s.io/v1beta3
certificatesDir: /etc/kubernetes/pki
clusterName: kubernetes
controllerManager: {}
dns: {}
etcd:
local:
dataDir: /var/lib/etcd
imageRepository: registry.k8s.io
kind: ClusterConfiguration
kubernetesVersion: 1.29.0
networking:
dnsDomain: cluster.local
serviceSubnet: 10.96.0.0/12
scheduler: {}
---
apiVersion: kubelet.config.k8s.io/v1beta1
authentication:
anonymous:
enabled: false
webhook:
cacheTTL: 0s
enabled: true
x509:
clientCAFile: /etc/kubernetes/pki/ca.crt
authorization:
mode: Webhook
webhook:
cacheAuthorizedTTL: 0s
cacheUnauthorizedTTL: 0s
cgroupDriver: systemd
clusterDNS:
- 10.96.0.10
clusterDomain: cluster.local
containerRuntimeEndpoint: ""
cpuManagerReconcilePeriod: 0s
evictionPressureTransitionPeriod: 0s
fileCheckFrequency: 0s
healthzBindAddress: 127.0.0.1
healthzPort: 10248
httpCheckFrequency: 0s
imageMaximumGCAge: 0s
imageMinimumGCAge: 0s
kind: KubeletConfiguration
logging:
flushFrequency: 0
options:
json:
infoBufferSize: "0"
verbosity: 0
memorySwap: {}
nodeStatusReportFrequency: 0s
nodeStatusUpdateFrequency: 0s
resolvConf: /run/systemd/resolve/resolv.conf
rotateCertificates: true
runtimeRequestTimeout: 0s
shutdownGracePeriod: 0s
shutdownGracePeriodCriticalPods: 0s
staticPodPath: /etc/kubernetes/manifests
streamingConnectionIdleTimeout: 0s
syncFrequency: 0s
volumeStatsAggPeriod: 0s从默认的配置中可以看到，可以使用imageRepository定制在集群初始化时拉取k8s所需镜像的地址。基于默认配置定制出本次使用kubeadm初始化集群所需的配置文件kubeadm.yaml：
apiVersion: kubeadm.k8s.io/v1beta3
kind: InitConfiguration
localAPIEndpoint:
advertiseAddress: 192.168.96.154
bindPort: 6443
nodeRegistration:
criSocket: unix:///run/containerd/containerd.sock
taints:
- effect: PreferNoSchedule
key: node-role.kubernetes.io/master
---
apiVersion: kubeadm.k8s.io/v1beta3
kind: ClusterConfiguration
kubernetesVersion: 1.29.0
imageRepository: registry.aliyuncs.com/google_containers
networking:
podSubnet: 10.244.0.0/16
---
apiVersion: kubelet.config.k8s.io/v1beta1
kind: KubeletConfiguration
cgroupDriver: systemd
failSwapOn: false
---
apiVersion: kubeproxy.config.k8s.io/v1alpha1
kind: KubeProxyConfiguration
mode: ipvs这里定制了imageRepository为阿里云的registry，避免因gcr被墙，无法直接拉取镜像。criSocket设置了容器运行时为containerd。
同时设置kubelet的cgroupDriver为systemd，设置kube-proxy代理模式为ipvs。
在开始初始化集群之前可以使用kubeadm config images pull --config kubeadm.yaml预先在各个服务器节点上拉取所k8s需要的容器镜像。
kubeadm config images list --config kubeadm.yaml
registry.aliyuncs.com/google_containers/kube-apiserver:v1.29.0
registry.aliyuncs.com/google_containers/kube-controller-manager:v1.29.0
registry.aliyuncs.com/google_containers/kube-scheduler:v1.29.0
registry.aliyuncs.com/google_containers/kube-proxy:v1.29.0
registry.aliyuncs.com/google_containers/coredns:v1.11.1
registry.aliyuncs.com/google_containers/pause:3.9
registry.aliyuncs.com/google_containers/etcd:3.5.10-0
kubeadm config images pull --config kubeadm.yaml
[config/images] Pulled registry.aliyuncs.com/google_containers/kube-apiserver:v1.29.0
[config/images] Pulled registry.aliyuncs.com/google_containers/kube-controller-manager:v1.29.0
[config/images] Pulled registry.aliyuncs.com/google_containers/kube-scheduler:v1.29.0
[config/images] Pulled registry.aliyuncs.com/google_containers/kube-proxy:v1.29.0
[config/images] Pulled registry.aliyuncs.com/google_containers/coredns:v1.11.1
[config/images] Pulled registry.aliyuncs.com/google_containers/pause:3.9
[config/images] Pulled registry.aliyuncs.com/google_containers/etcd:3.5.10-0接下来使用kubeadm初始化集群，选择node4作为Master Node，在node4上执行下面的命令：
kubeadm init --config kubeadm.yaml
[init] Using Kubernetes version: v1.29.0
[preflight] Running pre-flight checks
[WARNING Swap]: swap is supported for cgroup v2 only; the NodeSwap feature gate of the kubelet is beta but disabled by default
[preflight] Pulling images required for setting up a Kubernetes cluster
[preflight] This might take a minute or two, depending on the speed of your internet connection
[preflight] You can also perform this action in beforehand using 'kubeadm config images pull'
[certs] Using certificateDir folder "/etc/kubernetes/pki"
[certs] Using existing ca certificate authority
[certs] Using existing apiserver certificate and key on disk
[certs] Using existing apiserver-kubelet-client certificate and key on disk
[certs] Using existing front-proxy-ca certificate authority
[certs] Using existing front-proxy-client certificate and key on disk
[certs] Using existing etcd/ca certificate authority
[certs] Using existing etcd/server certificate and key on disk
[certs] Using existing etcd/peer certificate and key on disk
[certs] Using existing etcd/healthcheck-client certificate and key on disk
[certs] Using existing apiserver-etcd-client certificate and key on disk
[certs] Using the existing "sa" key
[kubeconfig] Using kubeconfig folder "/etc/kubernetes"
[kubeconfig] Writing "admin.conf" kubeconfig file
[kubeconfig] Writing "super-admin.conf" kubeconfig file
[kubeconfig] Writing "kubelet.conf" kubeconfig file
[kubeconfig] Writing "controller-manager.conf" kubeconfig file
[kubeconfig] Writing "scheduler.conf" kubeconfig file
[etcd] Creating static Pod manifest for local etcd in "/etc/kubernetes/manifests"
[control-plane] Using manifest folder "/etc/kubernetes/manifests"
[control-plane] Creating static Pod manifest for "kube-apiserver"
[control-plane] Creating static Pod manifest for "kube-controller-manager"
[control-plane] Creating static Pod manifest for "kube-scheduler"
[kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env"
[kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml"
[kubelet-start] Starting the kubelet
[wait-control-plane] Waiting for the kubelet to boot up the control plane as static Pods from directory "/etc/kubernetes/manifests". This can take up to 4m0s
[apiclient] All control plane components are healthy after 12.006365 seconds
[upload-config] Storing the configuration used in ConfigMap "kubeadm-config" in the "kube-system" Namespace
[kubelet] Creating a ConfigMap "kubelet-config" in namespace kube-system with the configuration for the kubelets in the cluster
[upload-certs] Skipping phase. Please see --upload-certs
[mark-control-plane] Marking the node node4 as control-plane by adding the labels: [node-role.kubernetes.io/control-plane node.kubernetes.io/exclude-from-external-load-balancers]
[mark-control-plane] Marking the node node4 as control-plane by adding the taints [node-role.kubernetes.io/master:PreferNoSchedule]
[bootstrap-token] Using token: alhelp.oyjw8wk6zyw5b55p
[bootstrap-token] Configuring bootstrap tokens, cluster-info ConfigMap, RBAC Roles
[bootstrap-token] Configured RBAC rules to allow Node Bootstrap tokens to get nodes
[bootstrap-token] Configured RBAC rules to allow Node Bootstrap tokens to post CSRs in order for nodes to get long term certificate credentials
[bootstrap-token] Configured RBAC rules to allow the csrapprover controller automatically approve CSRs from a Node Bootstrap Token
[bootstrap-token] Configured RBAC rules to allow certificate rotation for all node client certificates in the cluster
[bootstrap-token] Creating the "cluster-info" ConfigMap in the "kube-public" namespace
[kubelet-finalize] Updating "/etc/kubernetes/kubelet.conf" to point to a rotatable kubelet client certificate and key
[addons] Applied essential addon: CoreDNS
[addons] Applied essential addon: kube-proxy
Your Kubernetes control-plane has initialized successfully!
To start using your cluster, you need to run the following as a regular user:
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
Alternatively, if you are the root user, you can run:
export KUBECONFIG=/etc/kubernetes/admin.conf
You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
https://kubernetes.io/docs/concepts/cluster-administration/addons/
Then you can join any number of worker nodes by running the following on each as root:
kubeadm join 192.168.96.154:6443 --token alhelp.oyjw8wk6zyw5b55p \
--discovery-token-ca-cert-hash sha256:402b5d2d29367ada9b8ee2b37bfb246a318cbfce71d9e38c9117701455714f3e上面记录了完成的初始化输出的内容，根据输出的内容基本上可以看出手动初始化安装一个Kubernetes集群所需要的关键步骤。 其中有以下关键内容：
[certs]生成相关的各种证书[kubeconfig]生成相关的kubeconfig文件[kubelet-start]生成kubelet的配置文件"/var/lib/kubelet/config.yaml"[control-plane]使用/etc/kubernetes/manifests目录中的yaml文件创建apiserver、controller-manager、scheduler的静态pod[bootstraptoken]生成token记录下来，后边使用kubeadm join往集群中添加节点时会用到[addons]安装基本插件:CoreDNS,kube-proxy- 下面的命令是配置常规用户如何使用kubectl访问集群：
mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudo chown $(id -u):$(id -g) $HOME/.kube/config - 最后给出了将另外2个节点加入集群的命令:
kubeadm join 192.168.96.154:6443 --token alhelp.oyjw8wk6zyw5b55p \
--discovery-token-ca-cert-hash sha256:402b5d2d29367ada9b8ee2b37bfb246a318cbfce71d9e38c9117701455714f3e查看一下集群状态，确认个组件都处于healthy状态，结果出现了错误:
kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
controller-manager Healthy ok
scheduler Healthy ok
etcd-0 Healthy ok
kubectl get --raw='/readyz?verbose'
...
readyz check passed
kubectl get --raw='/livez?verbose'
...
livez check passed集群初始化如果遇到问题，可以使用kubeadm reset命令进行清理。
2.3 安装包管理器helm 3
Helm是Kubernetes的包管理器，后续流程也将使用Helm安装Kubernetes的常用组件。 这里先在master节点node4上安装helm。
wget https://get.helm.sh/helm-v3.13.3-linux-amd64.tar.gz
tar -zxvf helm-v3.13.3-linux-amd64.tar.gz
install -m 755 linux-amd64/helm /usr/local/bin/helm执行helm list确认没有错误输出。
2.4 部署Pod Network组件Calico
选择calico作为k8s的Pod网络组件，下面使用helm在k8s集群中安装calico。
下载tigera-operator的helm chart:
wget https://github.com/projectcalico/calico/releases/download/v3.27.0/tigera-operator-v3.27.0.tgz查看这个chart的中可定制的配置:
helm show values tigera-operator-v3.27.0.tgz
# imagePullSecrets is a special helm field which, when specified, creates a secret
# containing the pull secret which is used to pull all images deployed by this helm chart and the resulting operator.
# this field is a map where the key is the desired secret name and the value is the contents of the imagePullSecret.
#
# Example: --set-file imagePullSecrets.gcr=./pull-secret.json
imagePullSecrets: {}
installation:
enabled: true
kubernetesProvider: ""
# imagePullSecrets are configured on all images deployed by the tigera-operator.
# secrets specified here must exist in the tigera-operator namespace; they won't be created by the operator or helm.
# imagePullSecrets are a slice of LocalObjectReferences, which is the same format they appear as on deployments.
#
# Example: --set installation.imagePullSecrets[0].name=my-existing-secret
imagePullSecrets: []
apiServer:
enabled: true
certs:
node:
key:
cert:
commonName:
typha:
key:
cert:
commonName:
caBundle:
# Resource requests and limits for the tigera/operator pod.
resources: {}
# Tolerations for the tigera/operator pod.
tolerations:
- effect: NoExecute
operator: Exists
- effect: NoSchedule
operator: Exists
# NodeSelector for the tigera/operator pod.
nodeSelector:
kubernetes.io/os: linux
# Custom annotations for the tigera/operator pod.
podAnnotations: {}
# Custom labels for the tigera/operator pod.
podLabels: {}
# Image and registry configuration for the tigera/operator pod.
tigeraOperator:
image: tigera/operator
version: v1.32.3
registry: quay.io
calicoctl:
image: docker.io/calico/ctl
tag: v3.27.0
kubeletVolumePluginPath: /var/lib/kubelet定制的values.yaml如下:
# 可针对上面的配置进行定制,例如calico的镜像改成从私有库拉取。
# 这里只是个人本地环境测试k8s新版本，因此只有下面几行配置
apiServer:
enabled: false
installation:
kubeletVolumePluginPath: None使用helm安装calico：
helm install calico tigera-operator-v3.27.0.tgz -n kube-system --create-namespace -f values.yaml等待并确认所有pod处于Running状态:
kubectl get pod -n kube-system | grep tigera-operator
tigera-operator-55585899bf-qkr84 1/1 Running 0 26s
kubectl get pods -n calico-system
NAME READY STATUS RESTARTS AGE
calico-kube-controllers-6784546df7-5dzld 1/1 Running 0 5m30s
calico-node-24px9 1/1 Running 0 5m30s
calico-typha-75854bc9c9-5zvrb 1/1 Running 0 5m31s
csi-node-driver-mttxs 2/2 Running 0 5m30s查看一下calico向k8s中添加的api资源:
kubectl api-resources | grep calico
bgpconfigurations crd.projectcalico.org/v1 false BGPConfiguration
bgpfilters crd.projectcalico.org/v1 false BGPFilter
bgppeers crd.projectcalico.org/v1 false BGPPeer
blockaffinities crd.projectcalico.org/v1 false BlockAffinity
caliconodestatuses crd.projectcalico.org/v1 false CalicoNodeStatus
clusterinformations crd.projectcalico.org/v1 false ClusterInformation
felixconfigurations crd.projectcalico.org/v1 false FelixConfiguration
globalnetworkpolicies crd.projectcalico.org/v1 false GlobalNetworkPolicy
globalnetworksets crd.projectcalico.org/v1 false GlobalNetworkSet
hostendpoints crd.projectcalico.org/v1 false HostEndpoint
ipamblocks crd.projectcalico.org/v1 false IPAMBlock
ipamconfigs crd.projectcalico.org/v1 false IPAMConfig
ipamhandles crd.projectcalico.org/v1 false IPAMHandle
ippools crd.projectcalico.org/v1 false IPPool
ipreservations crd.projectcalico.org/v1 false IPReservation
kubecontrollersconfigurations crd.projectcalico.org/v1 false KubeControllersConfiguration
networkpolicies crd.projectcalico.org/v1 true NetworkPolicy
networksets crd.projectcalico.org/v1 true NetworkSet这些api资源是属于calico的，因此不建议使用kubectl来管理，推荐按照calicoctl来管理这些api资源。 将calicoctl安装为kubectl的插件:
cd /usr/local/bin
curl -o kubectl-calico -O -L "https://github.com/projectcalico/calico/releases/download/v3.27.0/calicoctl-linux-amd64"
chmod +x kubectl-calico验证插件正常工作:
kubectl calico -h2.5 验证k8s DNS是否可用
kubectl run curl --image=radial/busyboxplus:curl -it
If you don't see a command prompt, try pressing enter.
[ root@curl:/ ]$进入后执行nslookup kubernetes.default确认解析正常:
nslookup kubernetes.default
Server: 10.96.0.10
Address 1: 10.96.0.10 kube-dns.kube-system.svc.cluster.local
Name: kubernetes.default
Address 1: 10.96.0.1 kubernetes.default.svc.cluster.local2.6 向Kubernetes集群中添加Node节点
下面将node5, node6添加到Kubernetes集群中，分别在node5, node6上执行:
kubeadm join 192.168.96.154:6443 --token alhelp.oyjw8wk6zyw5b55p \
--discovery-token-ca-cert-hash sha256:402b5d2d29367ada9b8ee2b37bfb246a318cbfce71d9e38c9117701455714f3enode5和node6加入集群时遇到了如下问题，调度到node5(openEuler 22.03系统)或node6(Rock Linux 8.8)的calico-typha Pod无法启动，并报下面的错误:
kubectl describe po calico-node-ht7cf -n calico-system
...
kubelet Failed to create pod sandbox: open /run/systemd/resolve/resolv.conf: no such file or directory而/run/systemd/resolve/resolv.conf 文件是由systemd-resolved服务管理的。Ubuntu 22.04上默认安装并启动了这个服务。openEuler 22.03上没有安装这个服务。Rocky Linux 8.8上默认安装但没有启动这个服务。
下面node5上安装并启动systemd-resolved:
yum install -y systemd-resolved
systemctl enable systemd-resolved --now
systemctl status systemd-resolved在node6上启动systemd-resolved:
systemctl enable systemd-resolved --now
systemctl status systemd-resolved之后3个节点上的calico相关pod全部启动正常:
kubectl get po -n calico-system -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
calico-kube-controllers-6784546df7-5dzld 1/1 Running 0 20m 10.244.3.67 node4 <none> <none>
calico-node-24px9 1/1 Running 0 20m 192.168.96.154 node4 <none> <none>
calico-node-ht7cf 1/1 Running 0 7m6s 192.168.96.155 node5 <none> <none>
calico-node-tzql5 1/1 Running 0 7m4s 192.168.96.156 node6 <none> <none>
calico-typha-75854bc9c9-5zvrb 1/1 Running 0 20m 192.168.96.154 node4 <none> <none>
calico-typha-75854bc9c9-l8bmh 1/1 Running 0 6m56s 192.168.96.155 node5 <none> <none>
csi-node-driver-2mk5s 2/2 Running 0 7m6s 10.244.33.129 node5 <none> <none>
csi-node-driver-9qrzx 2/2 Running 0 7m4s 10.244.139.1 node6 <none> <none>
csi-node-driver-mttxs 2/2 Running 0 20m 10.244.3.65 node4 <none> <none>在master节点上执行命令查看集群中的节点(需要等待新加入节点上的calico-node pod启动正常)：
kubectl get node
NAME STATUS ROLES AGE VERSION
node4 Ready control-plane 34m v1.29.0
node5 Ready <none> 7m31s v1.29.0
node6 Ready <none> 7m29s v1.29.03.Kubernetes常用组件部署
3.1 使用Helm部署ingress-nginx
为了便于将集群中的服务暴露到集群外部，需要使用Ingress。接下来使用Helm将ingress-nginx部署到Kubernetes上。 Nginx Ingress Controller被部署在Kubernetes的边缘节点上。
这里将node4(192.168.96.154)作为边缘节点，打上Label：
kubectl label node node4 node-role.kubernetes.io/edge=下载ingress-nginx的helm chart:
wget https://github.com/kubernetes/ingress-nginx/releases/download/helm-chart-4.9.0/ingress-nginx-4.9.0.tgz查看ingress-nginx-4.9.0.tgz这个chart的可定制配置:
helm show values ingress-nginx-4.9.0.tgz对values.yaml配置定制如下:
controller:
ingressClassResource:
name: nginx
enabled: true
default: true
controllerValue: "k8s.io/ingress-nginx"
admissionWebhooks:
enabled: false
replicaCount: 1
image:
# registry: registry.k8s.io
# image: ingress-nginx/controller
# tag: "v1.9.5"
registry: docker.io
image: unreachableg/registry.k8s.io_ingress-nginx_controller
tag: "v1.9.5"
digest: sha256:bdc54c3e73dcec374857456559ae5757e8920174483882b9e8ff1a9052f96a35
hostNetwork: true
nodeSelector:
node-role.kubernetes.io/edge: ''
affinity:
podAntiAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app
operator: In
values:
- nginx-ingress
- key: component
operator: In
values:
- controller
topologyKey: kubernetes.io/hostname
tolerations:
- key: node-role.kubernetes.io/master
operator: Exists
effect: NoSchedule
- key: node-role.kubernetes.io/master
operator: Exists
effect: PreferNoSchedulenginx ingress controller的副本数replicaCount为1，将被调度到node4这个边缘节点上。这里并没有指定nginx ingress controller service的externalIPs，而是通过hostNetwork: true设置nginx ingress controller使用宿主机网络。
因为registry.k8s.io被墙，这里替换成unreachableg/registry.k8s.io_ingress-nginx_controller提前拉取一下镜像:
crictl --runtime-endpoint=unix:///run/containerd/containerd.sock pull unreachableg/registry.k8s.io_ingress-nginx_controller:v1.9.5helm install ingress-nginx ingress-nginx-4.9.0.tgz --create-namespace -n ingress-nginx -f values.yamlkubectl get po -n ingress-nginx
NAME READY STATUS RESTARTS AGE
ingress-nginx-controller-6445445cb8-c4fh4 1/1 Running 0 74s测试访问http://192.168.96.154返回默认的nginx 404页，则部署完成。
3.2 使用Helm部署dashboard
先部署metrics-server：
wget https://github.com/kubernetes-sigs/metrics-server/releases/download/v0.6.4/components.yaml修改components.yaml中的image为docker.io/unreachableg/k8s.gcr.io_metrics-server_metrics-server:v0.6.4。
修改components.yaml中容器的启动参数，加入--kubelet-insecure-tls。
kubectl apply -f components.yamlmetrics-server的pod正常启动后，等一段时间就可以使用kubectl top查看集群和pod的metrics信息:
kubectl top node
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
node4 373m 9% 2184Mi 27%
node5 42m 1% 968Mi 12%
node6 131m 3% 918Mi 12%
kubectl top pod -n kube-system
NAME CPU(cores) MEMORY(bytes)
coredns-857d9ff4c9-5pvft 6m 13Mi
coredns-857d9ff4c9-zkmm6 5m 13Mi
etcd-node4 62m 58Mi
kube-apiserver-node4 154m 350Mi
kube-controller-manager-node4 35m 45Mi
kube-proxy-4qfvt 31m 19Mi
kube-proxy-98k25 9m 17Mi
kube-proxy-rbh22 9m 18Mi
kube-scheduler-node4 9m 16Mi
metrics-server-7d686f4d9d-pxn8g 13m 18Mi
tigera-operator-55585899bf-qkr84 5m 28Mi接下来使用helm部署k8s的dashboard。当前k8s dashboard已经更新到了v3.0.0-alpha0，这里体验一下v3版本。
从k8s dashboard的v3版本开始，底层架构已更改，需要进行干净的安装，如果是在做升级dashboard操作，请首先移除先前的安装，这里是全新安装可以忽略。
k8s dashboard的v3版本现在默认使用cert-manager和nginx-ingress-controller。如果选择基于yaml清单的安装，请确保在集群中已安装它们。
我们前面已经安装了nginx-ingress-controller，下面先安装cert-manager:
wget https://github.com/cert-manager/cert-manager/releases/download/v1.13.3/cert-manager.yaml
kubectl apply -f cert-manager.yaml确保cert-manager的所有pod启动正常:
kubectl get po -n cert-manager
NAME READY STATUS RESTARTS AGE
cert-manager-6774cd657f-q9qpf 1/1 Running 0 102s
cert-manager-cainjector-55c8b7b49b-vf8r4 1/1 Running 0 102s
cert-manager-webhook-57797c469d-cgw4n 1/1 Running 0 102s下载dashboard的yaml清单文件:
wget https://raw.githubusercontent.com/kubernetes/dashboard/v3.0.0-alpha0/charts/kubernetes-dashboard.yaml编辑kubernetes-dashboard.yaml清单文件，将其中的ingress中的host替换想分配给k8s dashboard的域名:
kind: Ingress
apiVersion: networking.k8s.io/v1
metadata:
name: kubernetes-dashboard
namespace: kubernetes-dashboard
labels:
app.kubernetes.io/name: nginx-ingress
app.kubernetes.io/part-of: kubernetes-dashboard
annotations:
nginx.ingress.kubernetes.io/ssl-redirect: "true"
cert-manager.io/issuer: selfsigned
spec:
ingressClassName: nginx
tls:
- hosts:
- localhost
secretName: kubernetes-dashboard-certs
rules:
- host: k8s.example.com
http:
paths:
- path: /
pathType: Prefix
backend:
service:
name: kubernetes-dashboard-web
port:
name: web
- path: /api
pathType: Prefix
backend:
service:
name: kubernetes-dashboard-api
port:
name: api这里将k8s以ingress暴露到k8s集群外边，这里模拟真实环境中以域名形式访问暴露的服务。k8s dashboard的ingress中配置的域名为k8s.example.com，实际上需要客户端浏览器电脑的DNS可以解析到ingress controller(这里是192.168.96.154)上。
如果没有DNS，可以手动在客户端浏览器电脑设置hosts配置192.168.96.154 k8s.example.com。
安装dashboard的yaml清单文件:
kubectl apply -f kubernetes-dashboard.yaml确认dashboard的相关Pod启动正常:
kubectl get po -n kubernetes-dashboard
NAME READY STATUS RESTARTS AGE
kubernetes-dashboard-api-8586787f7-txzdx 1/1 Running 0 3m40s
kubernetes-dashboard-metrics-scraper-6959b784dc-424p5 1/1 Running 0 3m40s
kubernetes-dashboard-web-6b6d549b4-jcp2l 1/1 Running 0 3m40s
kubectl get ingress -n kubernetes-dashboard
NAME CLASS HOSTS ADDRESS PORTS AGE
kubernetes-dashboard nginx k8s.example.com 80, 443 3m49s创建管理员sa:
kubectl create serviceaccount kube-dashboard-admin-sa -n kube-system
kubectl create clusterrolebinding kube-dashboard-admin-sa \
--clusterrole=cluster-admin --serviceaccount=kube-system:kube-dashboard-admin-sa创建集群管理员登录dashboard所需token:
kubectl create token kube-dashboard-admin-sa -n kube-system --duration=87600h
eyJhbGciOiJSUzI1NiIsImtpZCI6Im5SWVpMcGZMcHFjYVdFcFNzX2kwTmwxYUx2M2NRckU5MFJBUmpSLW1fV28ifQ.eyJhdWQiOlsiaHR0cHM6Ly9rdWJlcm5ldGVzLmRlZmF1bHQuc3ZjLmNsdXN0ZXIubG9jYWwiXSwiZXhwIjoyMDE5MTExNjI5LCJpYXQiOjE3MDM3NTE2MjksImlzcyI6Imh0dHBzOi8va3ViZXJuZXRlcy5kZWZhdWx0LnN2Yy5jbHVzdGVyLmxvY2FsIiwia3ViZXJuZXRlcy5pbyI6eyJuYW1lc3BhY2UiOiJrdWJlLXN5c3RlbSIsInNlcnZpY2VhY2NvdW50Ijp7Im5hbWUiOiJrdWJlLWRhc2hib2FyZC1hZG1pbi1zYSIsInVpZCI6ImQ1YzZiMDdmLWUzMDAtNDMzOS04ZDY1LTUwYzg0N2FjMjg2MCJ9fSwibmJmIjoxNzAzNzUxNjI5LCJzdWIiOiJzeXN0ZW06c2VydmljZWFjY291bnQ6a3ViZS1zeXN0ZW06a3ViZS1kYXNoYm9hcmQtYWRtaW4tc2EifQ.EOFDNd0GvXjJpoUYFjOKDhuEbSJgLn6RuQeBgwjN-C4lR5C0URwXVarDUmGJTJZiAcHsajM1RGmR9u26vFvh9ZKTaQOkpJKYvJACiUwiOFZzGv_j2Cc5erZbiJskNMzl_Yt_fyACDpZpB20pjtT5e91C5Z7NPdgHbQsKt0Nkj6iLoIrGDihWBUEl33v1q1JixYyvtr9v2TcmmT8kQDmwluIsetW2TwN17ZVD1wsVz9iRgu0xwEWgzKh9FebQKJOsMmKWerca9ov_PD62ppElR0553-spgjjxow-rZ4mxn3u5M-dPfX57yIBQjczCd3jyEDedMs_RmRxUz_rtebdQAw使用上面的token登录k8s dashboard。
