当前位置：首页 > news >正文

Kubernetes 高可用集群安装

news 2025/11/1 7:30:52

一、节点规划

系统: CentOSmaster01     ip:192.168.1.20   内存：8G   CPU: 8C   master02     ip:192.168.1.21   内存：8G   CPU: 8Cmaster03     ip:192.168.1.22   内存：8G   CPU: 8Cnode01       ip:192.168.1.23   内存：8G   CPU: 8Cnode02       ip:192.168.1.24   内存：8G   CPU: 8Clb           ip:192.168.1.15Pod网段       10.96.0.0/12Service网段   172.16.0.1/12

二、版本选择

本次安装选择kubernetes 1.28 版本

三、hosts配置

1、master01配置

192.168.1.20 master01192.168.1.21 master02192.168.1.22 master03192.168.1.23 node01192.168.1.24 node02192.168.1.15 lb

2、同步至其他主机

for i in master02 master03 node01 node02;do   scp /etc/hosts root@$i:/etc/hosts;done

四、SSH配置

1、所有节点初始化SSH配置

ssh-keygen -t rsa

2、master01免密登录其他节点

for node in master01 master02 master03 node01 node02;do ssh-copy-id -i .ssh/id_rsa.pub root@$node;done;

五、修改yum源为aliyun

vim /etc/yum.repos.d/CentOS-Base.repo

[base]name=CentOS-$releasever - Base#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=os&infra=$infrabaseurl=http://mirrors.aliyun.com/centos/$releasever/os/$basearch/gpgcheck=1gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
#released updates [updates]name=CentOS-$releasever - Updates#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=updates&infra=$infrabaseurl=http://mirrors.aliyun.com/centos/$releasever/updates/$basearch/gpgcheck=1gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
#additional packages that may be useful[extras]name=CentOS-$releasever - Extras#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=extras&infra=$infrabaseurl=http://mirrors.aliyun.com/centos/$releasever/extras/$basearch/gpgcheck=1gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
#additional packages that extend functionality of existing packages[centosplus]name=CentOS-$releasever - Plus#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=centosplus&infra=$infrabaseurl=http://mirrors.aliyun.com/centos/$releasever/centosplus/$basearch/gpgcheck=1enabled=0gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

六、CentOS基本环境升级

1、安装基础工具

yum install wget jq psmisc vim net-tools telnet yum-utils device-mapper-persistent-data lvm2 git -y

2、所有节点关闭firewalld 、dnsmasq、selinux

systemctl disable --now firewalld systemctl disable --now dnsmasqsystemctl disable --now NetworkManager
setenforce 0sed -i 's#SELINUX=enforcing#SELINUX=disabled#g' /etc/sysconfig/selinuxsed -i 's#SELINUX=enforcing#SELINUX=disabled#g' /etc/selinux/config

3、所有节点关闭swap分区，fstab注释swap

swapoff -a && sysctl -w vm.swappiness=0sed -ri '/^[^#]*swap/s@^@#@' /etc/fstab

4、所有节点同步时间

4.1、安装ntpdate

rpm -ivh http://mirrors.wlnmp.com/centos/wlnmp-release-centos.noarch.rpmyum install ntpdate -y

4.2、同步时间

ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtimeecho 'Asia/Shanghai' >/etc/timezonentpdate time2.aliyun.com

crontab -e# 加入到crontab*/5 * * * * /usr/sbin/ntpdate time2.aliyun.com

5、所有节点配置limit

5.1、ulimit -SHn 65535

5.2、limits.conf

vim /etc/security/limits.conf

# 末尾添加如下内容* soft nofile 65536* hard nofile 131072* soft nproc 65535* hard nproc 655350* soft memlock unlimited* hard memlock unlimited

6、升级内核

6.1、下载

wget http://193.49.22.109/elrepo/kernel/el7/x86_64/RPMS/kernel-ml-devel-4.19.12-1.el7.elrepo.x86_64.rpmwget http://193.49.22.109/elrepo/kernel/el7/x86_64/RPMS/kernel-ml-4.19.12-1.el7.elrepo.x86_64.rpm

6.2、同步至其他节点

for node in master02 master03 node01 node02;do  scp kernel-ml* root@$node:/root/;done;

6.3、所有节点安装

cd /root && yum localinstall -y kernel-ml*

6.4、更改内核启动顺序

grub2-set-default  0 && grub2-mkconfig -o /etc/grub2.cfggrubby --args="user_namespace.enable=1" --update-kernel="$(grubby --default-kernel)"

6.5、检查默认内核是不是4.19

grubby --default-kernel

[root@master01 ~]# grubby --default-kernel/boot/vmlinuz-4.19.12-1.el7.elrepo.x86_64

6.6、所有节点重启，然后检查内核是不是4.19

uname -a

6.7、所有节点安装ipvsadm

yum install ipvsadm ipset sysstat conntrack libseccomp -y

6.7.1、所有节点配置ipvs模块

modprobe -- ip_vsmodprobe -- ip_vs_rrmodprobe -- ip_vs_wrrmodprobe -- ip_vs_shmodprobe -- nf_conntrack

6.7.2、ipvs.conf

vim /etc/modules-load.d/ipvs.conf

ip_vsip_vs_lcip_vs_wlcip_vs_rrip_vs_wrrip_vs_lblcip_vs_lblcrip_vs_dhip_vs_ship_vs_foip_vs_nqip_vs_sedip_vs_ftpip_vs_shnf_conntrackip_tablesip_setxt_setipt_setipt_rpfilteript_REJECTipip

然后执行

systemctl enable --now systemd-modules-load.service

6.7.3、检查是否加载

[root@master01 ~]#  lsmod | grep -e ip_vs -e nf_conntrackip_vs_ftp              16384  0 nf_nat                 32768  1 ip_vs_ftpip_vs_sed              16384  0 ip_vs_nq               16384  0 ip_vs_fo               16384  0 ip_vs_dh               16384  0 ip_vs_lblcr            16384  0 ip_vs_lblc             16384  0 ip_vs_wlc              16384  0 ip_vs_lc               16384  0 ip_vs_sh               16384  0 ip_vs_wrr              16384  0 ip_vs_rr               16384  0 ip_vs                 151552  24 ip_vs_wlc,ip_vs_rr,ip_vs_dh,ip_vs_lblcr,ip_vs_sh,ip_vs_fo,ip_vs_nq,ip_vs_lblc,ip_vs_wrr,ip_vs_lc,ip_vs_sed,ip_vs_ftpnf_conntrack          143360  2 nf_nat,ip_vsnf_defrag_ipv6         20480  1 nf_conntracknf_defrag_ipv4         16384  1 nf_conntracklibcrc32c              16384  4 nf_conntrack,nf_nat,xfs,ip_vs

7、内核参数配置

cat <<EOF > /etc/sysctl.d/k8s.confnet.ipv4.ip_forward = 1net.bridge.bridge-nf-call-iptables = 1net.bridge.bridge-nf-call-ip6tables = 1fs.may_detach_mounts = 1vm.overcommit_memory=1net.ipv4.conf.all.route_localnet = 1vm.panic_on_oom=0fs.inotify.max_user_watches=89100fs.file-max=52706963fs.nr_open=52706963net.netfilter.nf_conntrack_max=2310720net.ipv4.tcp_keepalive_time = 600net.ipv4.tcp_keepalive_probes = 3net.ipv4.tcp_keepalive_intvl =15net.ipv4.tcp_max_tw_buckets = 36000net.ipv4.tcp_tw_reuse = 1net.ipv4.tcp_max_orphans = 327680net.ipv4.tcp_orphan_retries = 3net.ipv4.tcp_syncookies = 1net.ipv4.tcp_max_syn_backlog = 16384net.ipv4.ip_conntrack_max = 65536net.ipv4.tcp_max_syn_backlog = 16384net.ipv4.tcp_timestamps = 0net.core.somaxconn = 16384EOFsysctl --system

8、所有节点配置完后，重启节点，保证服务重启后，配置依旧

lsmod | grep --color=auto -e ip_vs -e nf_conntrack

七、基本组件安装

1、Containerd作为Runtime

yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repoyum install docker-ce-20.10.* docker-ce-cli-20.10.* containerd -y

2、所有节点配置Containerd需要的模块

cat <<EOF | sudo tee /etc/modules-load.d/containerd.confoverlaybr_netfilterEOF
modprobe -- overlaymodprobe -- br_netfilter

3、所有节点配置Containerd所需要的内核

cat <<EOF | sudo tee /etc/sysctl.d/99-kubernetes-cri.confnet.bridge.bridge-nf-call-iptables  = 1net.ipv4.ip_forward                 = 1net.bridge.bridge-nf-call-ip6tables = 1EOF
sysctl --system

4、所有节点配置Containerd的配置文件

mkdir -p /etc/containerdcontainerd config default | tee /etc/containerd/config.toml

5、所有节点将Containerd的Cgroup改为Systemd

vim /etc/containerd/config.toml

找到containerd.runtimes.runc.options，添加SystemdCgroup = true

6、所有节点将sandbox_image的Pause镜像

registry.cn-hangzhou.aliyuncs.com/google_containers/pause:3.6

镜像加速器配置：

[plugins."io.containerd.grpc.v1.cri".registry.mirrors]  [plugins."io.containerd.grpc.v1.cri".registry.mirrors."docker.io"]    endpoint = ["https://xxxxxxx.mirror.swr.myhuaweicloud.com"]

完整的配置文件

disabled_plugins = []imports = []oom_score = 0plugin_dir = ""required_plugins = []root = "/var/lib/containerd"state = "/run/containerd"temp = ""version = 2
[cgroup]  path = ""
[debug]  address = ""  format = ""  gid = 0  level = ""  uid = 0
[grpc]  address = "/run/containerd/containerd.sock"  gid = 0  max_recv_message_size = 16777216  max_send_message_size = 16777216  tcp_address = ""  tcp_tls_ca = ""  tcp_tls_cert = ""  tcp_tls_key = ""  uid = 0
[metrics]  address = ""  grpc_histogram = false
[plugins]
  [plugins."io.containerd.gc.v1.scheduler"]    deletion_threshold = 0    mutation_threshold = 100    pause_threshold = 0.02    schedule_delay = "0s"    startup_delay = "100ms"
  [plugins."io.containerd.grpc.v1.cri"]    device_ownership_from_security_context = false    disable_apparmor = false    disable_cgroup = false    disable_hugetlb_controller = true    disable_proc_mount = false    disable_tcp_service = true    drain_exec_sync_io_timeout = "0s"    enable_selinux = false    enable_tls_streaming = false    enable_unprivileged_icmp = false    enable_unprivileged_ports = false    ignore_deprecation_warnings = []    ignore_image_defined_volumes = false    max_concurrent_downloads = 3    max_container_log_line_size = 16384    netns_mounts_under_state_dir = false    restrict_oom_score_adj = false    sandbox_image = "registry.cn-hangzhou.aliyuncs.com/google_containers/pause:3.6"    selinux_category_range = 1024    stats_collect_period = 10    stream_idle_timeout = "4h0m0s"    stream_server_address = "127.0.0.1"    stream_server_port = "0"    systemd_cgroup = false    tolerate_missing_hugetlb_controller = true    unset_seccomp_profile = ""
    [plugins."io.containerd.grpc.v1.cri".cni]      bin_dir = "/opt/cni/bin"      conf_dir = "/etc/cni/net.d"      conf_template = ""      ip_pref = ""      max_conf_num = 1
    [plugins."io.containerd.grpc.v1.cri".containerd]      default_runtime_name = "runc"      disable_snapshot_annotations = true      discard_unpacked_layers = false      ignore_rdt_not_enabled_errors = false      no_pivot = false      snapshotter = "overlayfs"
      [plugins."io.containerd.grpc.v1.cri".containerd.default_runtime]        base_runtime_spec = ""        cni_conf_dir = ""        cni_max_conf_num = 0        container_annotations = []        pod_annotations = []        privileged_without_host_devices = false        runtime_engine = ""        runtime_path = ""        runtime_root = ""        runtime_type = ""
        [plugins."io.containerd.grpc.v1.cri".containerd.default_runtime.options]
      [plugins."io.containerd.grpc.v1.cri".containerd.runtimes]
        [plugins."io.containerd.grpc.v1.cri".containerd.runtimes.runc]          base_runtime_spec = ""          cni_conf_dir = ""          cni_max_conf_num = 0          container_annotations = []          pod_annotations = []          privileged_without_host_devices = false          runtime_engine = ""          runtime_path = ""          runtime_root = ""          runtime_type = "io.containerd.runc.v2"
          [plugins."io.containerd.grpc.v1.cri".containerd.runtimes.runc.options]            BinaryName = ""            CriuImagePath = ""            CriuPath = ""            CriuWorkPath = ""            IoGid = 0            IoUid = 0            NoNewKeyring = false            NoPivotRoot = false            Root = ""            ShimCgroup = ""            SystemdCgroup = true
      [plugins."io.containerd.grpc.v1.cri".containerd.untrusted_workload_runtime]        base_runtime_spec = ""        cni_conf_dir = ""        cni_max_conf_num = 0        container_annotations = []        pod_annotations = []        privileged_without_host_devices = false        runtime_engine = ""        runtime_path = ""        runtime_root = ""        runtime_type = ""
        [plugins."io.containerd.grpc.v1.cri".containerd.untrusted_workload_runtime.options]
    [plugins."io.containerd.grpc.v1.cri".image_decryption]      key_model = "node"
    [plugins."io.containerd.grpc.v1.cri".registry]      config_path = ""
      [plugins."io.containerd.grpc.v1.cri".registry.auths]
      [plugins."io.containerd.grpc.v1.cri".registry.configs]
      [plugins."io.containerd.grpc.v1.cri".registry.headers]
      [plugins."io.containerd.grpc.v1.cri".registry.mirrors]        [plugins."io.containerd.grpc.v1.cri".registry.mirrors."docker.io"]          endpoint = ["https://3a0ac22010f3428eb2cae4ab2aa014f5.mirror.swr.myhuaweicloud.com"]
    [plugins."io.containerd.grpc.v1.cri".x509_key_pair_streaming]      tls_cert_file = ""      tls_key_file = ""
  [plugins."io.containerd.internal.v1.opt"]    path = "/opt/containerd"
  [plugins."io.containerd.internal.v1.restart"]    interval = "10s"
  [plugins."io.containerd.internal.v1.tracing"]
  [plugins."io.containerd.metadata.v1.bolt"]    content_sharing_policy = "shared"
  [plugins."io.containerd.monitor.v1.cgroups"]    no_prometheus = false
  [plugins."io.containerd.runtime.v1.linux"]    no_shim = false    runtime = "runc"    runtime_root = ""    shim = "containerd-shim"    shim_debug = false
  [plugins."io.containerd.runtime.v2.task"]    platforms = ["linux/amd64"]    sched_core = false
  [plugins."io.containerd.service.v1.diff-service"]    default = ["walking"]
  [plugins."io.containerd.service.v1.tasks-service"]    rdt_config_file = ""
  [plugins."io.containerd.snapshotter.v1.aufs"]    root_path = ""
  [plugins."io.containerd.snapshotter.v1.btrfs"]    root_path = ""
  [plugins."io.containerd.snapshotter.v1.devmapper"]    async_remove = false    base_image_size = ""    discard_blocks = false    fs_options = ""    fs_type = ""    pool_name = ""    root_path = ""
  [plugins."io.containerd.snapshotter.v1.native"]    root_path = ""
  [plugins."io.containerd.snapshotter.v1.overlayfs"]    mount_options = []    root_path = ""    sync_remove = false    upperdir_label = false
  [plugins."io.containerd.snapshotter.v1.zfs"]    root_path = ""
  [plugins."io.containerd.tracing.processor.v1.otlp"]
[proxy_plugins]
[stream_processors]
  [stream_processors."io.containerd.ocicrypt.decoder.v1.tar"]    accepts = ["application/vnd.oci.image.layer.v1.tar+encrypted"]    args = ["--decryption-keys-path", "/etc/containerd/ocicrypt/keys"]    env = ["OCICRYPT_KEYPROVIDER_CONFIG=/etc/containerd/ocicrypt/ocicrypt_keyprovider.conf"]    path = "ctd-decoder"    returns = "application/vnd.oci.image.layer.v1.tar"
  [stream_processors."io.containerd.ocicrypt.decoder.v1.tar.gzip"]    accepts = ["application/vnd.oci.image.layer.v1.tar+gzip+encrypted"]    args = ["--decryption-keys-path", "/etc/containerd/ocicrypt/keys"]    env = ["OCICRYPT_KEYPROVIDER_CONFIG=/etc/containerd/ocicrypt/ocicrypt_keyprovider.conf"]    path = "ctd-decoder"    returns = "application/vnd.oci.image.layer.v1.tar+gzip"
[timeouts]  "io.containerd.timeout.bolt.open" = "0s"  "io.containerd.timeout.shim.cleanup" = "5s"  "io.containerd.timeout.shim.load" = "5s"  "io.containerd.timeout.shim.shutdown" = "3s"  "io.containerd.timeout.task.state" = "2s"
[ttrpc]  address = ""  gid = 0  uid = 0

7、所有节点启动Containerd，并配置开机自启动

systemctl daemon-reloadsystemctl enable --now containerd

8、所有节点配置crictl客户端连接的运行时位置

cat > /etc/crictl.yaml <<EOFruntime-endpoint: unix:///run/containerd/containerd.sockimage-endpoint: unix:///run/containerd/containerd.socktimeout: 10debug: falseEOF

八、ETCD安装

1、下载

 wget https://github.com/etcd-io/etcd/releases/download/v3.5.7/etcd-v3.5.7-linux-amd64.tar.gz

2、解压ETCD安装包，安装至执行目录

tar -zxvf etcd-v3.5.7-linux-amd64.tar.gz --strip-components=1 -C /usr/local/bin etcd-v3.5.7-linux-amd64/etcd{,ctl}

3、同步执行文件至master02、master03节点

for node in master02 master03;do   scp /usr/local/bin/etcd* root@$node:/usr/local/bin/;done;

九、K8S安装

1、下载

wget https://dl.k8s.io/v1.28.15/kubernetes-server-linux-amd64.tar.gz

2、解压安装包至执行目录

 tar -zvxf kubernetes-server-linux-amd64.tar.gz  --strip-components=3 -C /usr/local/bin kubernetes/server/bin/kube{let,ctl,-apiserver,-controller-manager,-scheduler,-proxy}

3、同步执行文件到其他节点

MasterNodes='master02 master03'WorkNodes='node01 node02'for node in $MasterNodes;do   scp /usr/local/bin/kube{let,ctl,-apiserver,-controller-manager,-scheduler,-proxy} root@$node:/usr/local/bin/;done;
for node in $WorkNodes;do  scp /usr/local/bin/kube{let,-proxy} root@$node:/usr/local/bin/;done;

十、生成证书

1、下载生成证书工具

在master01操作

wget "https://pkg.cfssl.org/R1.2/cfssl_linux-amd64" -O /usr/local/bin/cfsslwget "https://pkg.cfssl.org/R1.2/cfssljson_linux-amd64" -O /usr/local/bin/cfssljsonchmod +x /usr/local/bin/cfssl /usr/local/bin/cfssljson

2、ETCD证书

2.1、所有Master节点创建etcd证书目录

mkdir /etc/etcd/ssl -p

2.2、clone相关配置文件

git clone https://gitee.com/dukuan/k8s-ha-install.git# 切换至v1.28版本git checkout manual-installation-v1.28.x# 切换至cd /root/k8s-ha-install/pki

2.3、生成ETCD CA证书和CA证书的key

cfssl gencert -initca etcd-ca-csr.json | cfssljson -bare /etc/etcd/ssl/etcd-cacfssl gencert \-ca=/etc/etcd/ssl/etcd-ca.pem \-ca-key=/etc/etcd/ssl/etcd-ca-key.pem \-config=ca-config.json \-hostname=127.0.0.1,master01,master02,master03,192.168.1.20,192.168.1.21,192.168.1.22 \   -profile=kubernetes \etcd-csr.json | cfssljson -bare /etc/etcd/ssl/etcd

2.4、复制证书至其他master节点

MasterNodes='master02 master03'for NODE in $MasterNodes; do  ssh root@$NODE "mkdir -p /etc/etcd/ssl"for FILE in etcd-ca-key.pem  etcd-ca.pem  etcd-key.pem  etcd.pem; do  scp /etc/etcd/ssl/${FILE} root@$NODE:/etc/etcd/ssl/${FILE}donedone

3、K8S组件证书

3.1、Master01生成kubernetes证书

cd /root/k8s-ha-install/pkimkdir -p /etc/kubernetes/pkicfssl gencert -initca ca-csr.json | cfssljson -bare /etc/kubernetes/pki/ca

#172.16.0.1是k8sservice的网段，如果说需要更改k8sservice网段，那就需要更改172.16.0.1，

# 如果不是高可用集群，192.168.1.15为Master01的IP

cfssl gencert   \  -ca=/etc/kubernetes/pki/ca.pem   \  -ca-key=/etc/kubernetes/pki/ca-key.pem   \  -config=ca-config.json   \  -hostname=172.16.0.1,192.168.1.15,127.0.0.1,kubernetes,kubernetes.default,kubernetes.default.svc,kubernetes.default.svc.cluster,kubernetes.default.svc.cluster.local,192.168.1.20,192.168.1.21,192.168.1.22   \  -profile=kubernetes   apiserver-csr.json | cfssljson -bare /etc/kubernetes/pki/apiserver

3.2、生成apiserver的聚合证书

cfssl gencert   -initca front-proxy-ca-csr.json | cfssljson -bare /etc/kubernetes/pki/front-proxy-ca cfssl gencert   -ca=/etc/kubernetes/pki/front-proxy-ca.pem   -ca-key=/etc/kubernetes/pki/front-proxy-ca-key.pem   -config=ca-config.json   -profile=kubernetes   front-proxy-client-csr.json | cfssljson -bare /etc/kubernetes/pki/front-proxy-client

忽略这个warning

2025/03/30 00:16:58 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable forwebsites. For more information see the Baseline Requirements for the Issuance and Managementof Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org);specifically, section 10.2.3 ("Information Requirements")

3.3、生成controller-manage的证书

cfssl gencert \-ca=/etc/kubernetes/pki/ca.pem \-ca-key=/etc/kubernetes/pki/ca-key.pem \-config=ca-config.json \-profile=kubernetes \manager-csr.json | cfssljson -bare /etc/kubernetes/pki/controller-manager

3.4、set-cluster：设置一个集群项

# 注意，如果不是高可用集群，192.168.1.15:8443改为master01的地址，8443改为apiserver的端口，默认是6443

kubectl config set-cluster kubernetes \--certificate-authority=/etc/kubernetes/pki/ca.pem \--embed-certs=true \--server=https://192.168.1.15:8443 \--kubeconfig=/etc/kubernetes/controller-manager.kubeconfig

3.5、设置一个环境项，一个上下文

kubectl config set-context system:kube-controller-manager@kubernetes \--cluster=kubernetes \--user=system:kube-controller-manager \--kubeconfig=/etc/kubernetes/controller-manager.kubeconfig

3.6、set-credentials 设置一个用户项

kubectl config set-credentials system:kube-controller-manager \--client-certificate=/etc/kubernetes/pki/controller-manager.pem \--client-key=/etc/kubernetes/pki/controller-manager-key.pem \--embed-certs=true \--kubeconfig=/etc/kubernetes/controller-manager.kubeconfig

3.7、使用某个环境当做默认环境

kubectl config use-context system:kube-controller-manager@kubernetes \--kubeconfig=/etc/kubernetes/controller-manager.kubeconfig

3.8、生成kube-scheduler证书

cfssl gencert \-ca=/etc/kubernetes/pki/ca.pem \-ca-key=/etc/kubernetes/pki/ca-key.pem \-config=ca-config.json \-profile=kubernetes \scheduler-csr.json | cfssljson -bare /etc/kubernetes/pki/scheduler

# 注意，如果不是高可用集群，192.168.1.15:8443改为master01的地址，8443改为apiserver的端口，默认是6443

kubectl config set-cluster kubernetes \  --certificate-authority=/etc/kubernetes/pki/ca.pem \  --embed-certs=true \  --server=https://192.168.1.15:8443 \  --kubeconfig=/etc/kubernetes/scheduler.kubeconfig
kubectl config set-credentials system:kube-scheduler \--client-certificate=/etc/kubernetes/pki/scheduler.pem \--client-key=/etc/kubernetes/pki/scheduler-key.pem \--embed-certs=true \--kubeconfig=/etc/kubernetes/scheduler.kubeconfig
kubectl config set-context system:kube-scheduler@kubernetes \--cluster=kubernetes \--user=system:kube-scheduler \--kubeconfig=/etc/kubernetes/scheduler.kubeconfigkubectl config use-context system:kube-scheduler@kubernetes \--kubeconfig=/etc/kubernetes/scheduler.kubeconfig

3.9、配置admin证书

cfssl gencert \-ca=/etc/kubernetes/pki/ca.pem \-ca-key=/etc/kubernetes/pki/ca-key.pem \-config=ca-config.json \-profile=kubernetes \admin-csr.json | cfssljson -bare /etc/kubernetes/pki/admin

# 注意，如果不是高可用集群，192.168.1.15:8443改为master01的地址，8443改为apiserver的端口，默认是6443

kubectl config set-cluster kubernetes     \  --certificate-authority=/etc/kubernetes/pki/ca.pem     \  --embed-certs=true     \  --server=https://192.168.1.15:8443     \  --kubeconfig=/etc/kubernetes/admin.kubeconfig
kubectl config set-credentials kubernetes-admin     \  --client-certificate=/etc/kubernetes/pki/admin.pem     \  --client-key=/etc/kubernetes/pki/admin-key.pem     \  --embed-certs=true     \  --kubeconfig=/etc/kubernetes/admin.kubeconfig
kubectl config set-context kubernetes-admin@kubernetes     \  --cluster=kubernetes     \  --user=kubernetes-admin     \  --kubeconfig=/etc/kubernetes/admin.kubeconfig
kubectl config use-context kubernetes-admin@kubernetes     \  --kubeconfig=/etc/kubernetes/admin.kubeconfig

3.10、创建ServiceAccount Key -> secret

openssl genrsa -out /etc/kubernetes/pki/sa.key 2048
openssl rsa -in /etc/kubernetes/pki/sa.key -pubout -out /etc/kubernetes/pki/sa.pub

4、将证书发送至其他节点

for NODE in master02 master03; do for FILE in $(ls /etc/kubernetes/pki | grep -v etcd); do   scp /etc/kubernetes/pki/${FILE} $NODE:/etc/kubernetes/pki/${FILE};done; for FILE in admin.kubeconfig controller-manager.kubeconfig scheduler.kubeconfig; do   scp /etc/kubernetes/${FILE} $NODE:/etc/kubernetes/${FILE};done;done

查看证书文件

ls /etc/kubernetes/pki/
ls /etc/kubernetes/pki/ | wc -l

十一、ETCD配置

1、master01配置

vim /etc/etcd/etcd.config.yml

name: 'master01'data-dir: /var/lib/etcdwal-dir: /var/lib/etcd/walsnapshot-count: 5000heartbeat-interval: 100election-timeout: 1000quota-backend-bytes: 0listen-peer-urls: 'https://192.168.1.20:2380'listen-client-urls: 'https://192.168.1.20:2379,http://127.0.0.1:2379'max-snapshots: 3max-wals: 5cors:initial-advertise-peer-urls: 'https://192.168.1.20:2380'advertise-client-urls: 'https://192.168.1.20:2379'discovery:discovery-fallback: 'proxy'discovery-proxy:discovery-srv:initial-cluster: 'master01=https://192.168.1.20:2380,master02=https://192.168.1.21:2380,master03=https://192.168.1.22:2380'initial-cluster-token: 'etcd-k8s-cluster'initial-cluster-state: 'new'strict-reconfig-check: falseenable-v2: trueenable-pprof: trueproxy: 'off'proxy-failure-wait: 5000proxy-refresh-interval: 30000proxy-dial-timeout: 1000proxy-write-timeout: 5000proxy-read-timeout: 0client-transport-security:  cert-file: '/etc/kubernetes/pki/etcd/etcd.pem'  key-file: '/etc/kubernetes/pki/etcd/etcd-key.pem'  client-cert-auth: true  trusted-ca-file: '/etc/kubernetes/pki/etcd/etcd-ca.pem'  auto-tls: truepeer-transport-security:  cert-file: '/etc/kubernetes/pki/etcd/etcd.pem'  key-file: '/etc/kubernetes/pki/etcd/etcd-key.pem'  peer-client-cert-auth: true  trusted-ca-file: '/etc/kubernetes/pki/etcd/etcd-ca.pem'  auto-tls: truedebug: falselog-package-levels:log-outputs: [default]force-new-cluster: false

2、master02配置

vim /etc/etcd/etcd.config.yml

name: 'master02'data-dir: /var/lib/etcdwal-dir: /var/lib/etcd/walsnapshot-count: 5000heartbeat-interval: 100election-timeout: 1000quota-backend-bytes: 0listen-peer-urls: 'https://192.168.1.21:2380'listen-client-urls: 'https://192.168.1.21:2379,http://127.0.0.1:2379'max-snapshots: 3max-wals: 5cors:initial-advertise-peer-urls: 'https://192.168.1.21:2380'advertise-client-urls: 'https://192.168.1.21:2379'discovery:discovery-fallback: 'proxy'discovery-proxy:discovery-srv:initial-cluster: 'master01=https://192.168.1.20:2380,master02=https://192.168.1.21:2380,master03=https://192.168.1.22:2380'initial-cluster-token: 'etcd-k8s-cluster'initial-cluster-state: 'new'strict-reconfig-check: falseenable-v2: trueenable-pprof: trueproxy: 'off'proxy-failure-wait: 5000proxy-refresh-interval: 30000proxy-dial-timeout: 1000proxy-write-timeout: 5000proxy-read-timeout: 0client-transport-security:  cert-file: '/etc/kubernetes/pki/etcd/etcd.pem'  key-file: '/etc/kubernetes/pki/etcd/etcd-key.pem'  client-cert-auth: true  trusted-ca-file: '/etc/kubernetes/pki/etcd/etcd-ca.pem'  auto-tls: truepeer-transport-security:  cert-file: '/etc/kubernetes/pki/etcd/etcd.pem'  key-file: '/etc/kubernetes/pki/etcd/etcd-key.pem'  peer-client-cert-auth: true  trusted-ca-file: '/etc/kubernetes/pki/etcd/etcd-ca.pem'  auto-tls: truedebug: falselog-package-levels:log-outputs: [default]force-new-cluster: false

3、master03配置

vim /etc/etcd/etcd.config.yml

name: 'master03'data-dir: /var/lib/etcdwal-dir: /var/lib/etcd/walsnapshot-count: 5000heartbeat-interval: 100election-timeout: 1000quota-backend-bytes: 0listen-peer-urls: 'https://192.168.1.22:2380'listen-client-urls: 'https://192.168.1.22:2379,http://127.0.0.1:2379'max-snapshots: 3max-wals: 5cors:initial-advertise-peer-urls: 'https://192.168.1.22:2380'advertise-client-urls: 'https://192.168.1.22:2379'discovery:discovery-fallback: 'proxy'discovery-proxy:discovery-srv:initial-cluster: 'master01=https://192.168.1.20:2380,master02=https://192.168.1.21:2380,master03=https://192.168.1.22:2380'initial-cluster-token: 'etcd-k8s-cluster'initial-cluster-state: 'new'strict-reconfig-check: falseenable-v2: trueenable-pprof: trueproxy: 'off'proxy-failure-wait: 5000proxy-refresh-interval: 30000proxy-dial-timeout: 1000proxy-write-timeout: 5000proxy-read-timeout: 0client-transport-security:  cert-file: '/etc/kubernetes/pki/etcd/etcd.pem'  key-file: '/etc/kubernetes/pki/etcd/etcd-key.pem'  client-cert-auth: true  trusted-ca-file: '/etc/kubernetes/pki/etcd/etcd-ca.pem'  auto-tls: truepeer-transport-security:  cert-file: '/etc/kubernetes/pki/etcd/etcd.pem'  key-file: '/etc/kubernetes/pki/etcd/etcd-key.pem'  peer-client-cert-auth: true  trusted-ca-file: '/etc/kubernetes/pki/etcd/etcd-ca.pem'  auto-tls: truedebug: falselog-package-levels:log-outputs: [default]force-new-cluster: false

4、创建service

所有Master节点创建etcd service并启动

vim /usr/lib/systemd/system/etcd.service

[Unit]Description=Etcd ServiceDocumentation=https://coreos.com/etcd/docs/latest/After=network.target
[Service]Type=notifyExecStart=/usr/local/bin/etcd --config-file=/etc/etcd/etcd.config.ymlRestart=on-failureRestartSec=10LimitNOFILE=65536
[Install]WantedBy=multi-user.targetAlias=etcd3.service

所有Master节点创建etcd的证书目录

mkdir /etc/kubernetes/pki/etcdln -s /etc/etcd/ssl/* /etc/kubernetes/pki/etcd/systemctl daemon-reloadsystemctl enable --now etcd

5、查看ETCD的状态

export ETCDCTL_API=3etcdctl \  --endpoints="192.168.1.20:2379,192.168.1.21:2379,192.168.1.22:2379" \  --cacert=/etc/kubernetes/pki/etcd/etcd-ca.pem \  --cert=/etc/kubernetes/pki/etcd/etcd.pem \  --key=/etc/kubernetes/pki/etcd/etcd-key.pem  endpoint status --write-out=table

十二、高可用配置

1、所有Master节点安装keepalived和haproxy

yum install keepalived haproxy -y

2、所有Master配置HAProxy，配置一样

vim /etc/haproxy/haproxy.cfg

globalmaxconn  2000ulimit-n  16384log  127.0.0.1 local0 errstats timeout 30s
defaultslog globalmode  httpoption  httplogtimeout connect 5000timeout client  50000timeout server  50000timeout http-request 15stimeout http-keep-alive 15s
frontend k8smasterbind 0.0.0.0:8443bind 127.0.0.1:8443mode tcpoption tcplogtcp-request inspect-delay 5sdefault_backend k8smaster
backend k8smastermode tcpoption tcplogoption tcp-checkbalance roundrobindefault-server inter 10s downinter 5s rise 2 fall 2 slowstart 60s maxconn 250 maxqueue 256 weight 100server master01    192.168.1.20:6443  checkserver master02    192.168.1.21:6443  checkserver master03    192.168.1.22:6443  check

3、Master01 keepalived

所有Master节点配置KeepAlived，配置不一样，注意区分

[root@master01 pki]# vim /etc/keepalived/keepalived.conf ，注意每个节点的IP和网卡（interface参数）

! Configuration File for keepalivedglobal_defs {router_id LVS_DEVEL}vrrp_script chk_apiserver {script "/etc/keepalived/check_apiserver.sh"interval 5 weight -5fall 2rise 1}vrrp_instance VI_1 {state MASTERinterface ens192mcast_src_ip 192.168.1.20virtual_router_id 51priority 101   nopreemptadvert_int 2authentication {auth_type PASSauth_pass K8SHA_KA_AUTH}virtual_ipaddress {192.168.1.15}track_script {      chk_apiserver } }

4、Master02 keepalived

! Configuration File for keepalivedglobal_defs {router_id LVS_DEVEL}vrrp_script chk_apiserver {script "/etc/keepalived/check_apiserver.sh"interval 5 weight -5fall 2rise 1
}vrrp_instance VI_1 {state BACKUPinterface ens192mcast_src_ip 192.168.1.21virtual_router_id 51priority 100nopreemptadvert_int 2authentication {auth_type PASSauth_pass K8SHA_KA_AUTH}virtual_ipaddress {192.168.1.15}track_script {      chk_apiserver } }

5、Master03 keepalived

! Configuration File for keepalivedglobal_defs {router_id LVS_DEVEL}vrrp_script chk_apiserver {script "/etc/keepalived/check_apiserver.sh"interval 5weight -5fall 2     rise 1}vrrp_instance VI_1 {state BACKUPinterface ens192mcast_src_ip 192.168.1.22virtual_router_id 51  priority 100nopreemptadvert_int 2authentication {auth_type PASSauth_pass K8SHA_KA_AUTH}virtual_ipaddress {192.168.1.15}track_script {      chk_apiserver } }

6、健康检测配置

所有master节点

# vim /etc/keepalived/check_apiserver.sh 
#!/bin/bash
err=0for k in $(seq 1 3)do    check_code=$(pgrep haproxy)    if [[ $check_code == "" ]]; then        err=$(expr $err + 1)        sleep 1        continue    else        err=0        break    fidone
if [[ $err != "0" ]]; then    echo "systemctl stop keepalived"    /usr/bin/systemctl stop keepalived    exit 1else    exit 0fi

chmod +x /etc/keepalived/check_apiserver.sh

7、所有master节点启动haproxy和keepalived

systemctl daemon-reloadsystemctl enable --now haproxysystemctl enable --now keepalived

8、VIP测试

ping 192.168.1.15

重要：如果安装了keepalived和haproxy，需要测试keepalived是否是正常的

telnet 192.168.1.15 8443

十三、Kubernetes组件配置

所有节点创建相关目录

mkdir -p /etc/kubernetes/manifests/ /etc/systemd/system/kubelet.service.d /var/lib/kubelet /var/log/kubernetes

1、Apiserver

所有Master节点创建kube-apiserver service，# 注意，如果不是高可用集群，192.168.1.15改为master01的地址

1.1、Master01配置

注意本文档使用的k8s service网段为172.16.0.1/16，该网段不能和宿主机的网段、Pod网段的重复，请按需修改

# vim /usr/lib/systemd/system/kube-apiserver.service [Unit]Description=Kubernetes API ServerDocumentation=https://github.com/kubernetes/kubernetesAfter=network.target
[Service]ExecStart=/usr/local/bin/kube-apiserver \--v=2  \--logging-format=text  \--allow-privileged=true  \--bind-address=0.0.0.0  \--secure-port=6443  \--advertise-address=192.168.1.20 \--service-cluster-ip-range=172.16.0.1/12  \--service-node-port-range=10000-40000  \--etcd-servers=https://192.168.1.20:2379,https://192.168.1.21:2379,https://192.168.1.22:2379 \--etcd-cafile=/etc/etcd/ssl/etcd-ca.pem  \--etcd-certfile=/etc/etcd/ssl/etcd.pem  \--etcd-keyfile=/etc/etcd/ssl/etcd-key.pem  \--client-ca-file=/etc/kubernetes/pki/ca.pem  \--tls-cert-file=/etc/kubernetes/pki/apiserver.pem  \--tls-private-key-file=/etc/kubernetes/pki/apiserver-key.pem  \--kubelet-client-certificate=/etc/kubernetes/pki/apiserver.pem  \--kubelet-client-key=/etc/kubernetes/pki/apiserver-key.pem  \--service-account-key-file=/etc/kubernetes/pki/sa.pub  \--service-account-signing-key-file=/etc/kubernetes/pki/sa.key  \--service-account-issuer=https://kubernetes.default.svc.cluster.local \--kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname  \--enable-admission-plugins=NamespaceLifecycle,LimitRanger,ServiceAccount,DefaultStorageClass,DefaultTolerationSeconds,NodeRestriction,ResourceQuota  \--authorization-mode=Node,RBAC  \--enable-bootstrap-token-auth=true  \--requestheader-client-ca-file=/etc/kubernetes/pki/front-proxy-ca.pem  \--proxy-client-cert-file=/etc/kubernetes/pki/front-proxy-client.pem  \--proxy-client-key-file=/etc/kubernetes/pki/front-proxy-client-key.pem  \--requestheader-allowed-names=aggregator  \--requestheader-group-headers=X-Remote-Group  \--requestheader-extra-headers-prefix=X-Remote-Extra-  \--requestheader-username-headers=X-Remote-User# --token-auth-file=/etc/kubernetes/token.csv
Restart=on-failureRestartSec=10sLimitNOFILE=65535
[Install]WantedBy=multi-user.target

1.2、Master02配置

[Unit]Description=Kubernetes API ServerDocumentation=https://github.com/kubernetes/kubernetesAfter=network.target
[Service]ExecStart=/usr/local/bin/kube-apiserver \--v=2  \--logging-format=text  \--allow-privileged=true  \--bind-address=0.0.0.0  \--secure-port=6443  \--advertise-address=192.168.1.21 \--service-cluster-ip-range=172.16.0.1/12  \--service-node-port-range=10000-40000  \--etcd-servers=https://192.168.1.20:2379,https://192.168.1.21:2379,https://192.168.1.22:2379 \--etcd-cafile=/etc/etcd/ssl/etcd-ca.pem  \--etcd-certfile=/etc/etcd/ssl/etcd.pem  \--etcd-keyfile=/etc/etcd/ssl/etcd-key.pem  \--client-ca-file=/etc/kubernetes/pki/ca.pem  \--tls-cert-file=/etc/kubernetes/pki/apiserver.pem  \--tls-private-key-file=/etc/kubernetes/pki/apiserver-key.pem  \--kubelet-client-certificate=/etc/kubernetes/pki/apiserver.pem  \--kubelet-client-key=/etc/kubernetes/pki/apiserver-key.pem  \--service-account-key-file=/etc/kubernetes/pki/sa.pub  \--service-account-signing-key-file=/etc/kubernetes/pki/sa.key  \--service-account-issuer=https://kubernetes.default.svc.cluster.local \--kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname  \--enable-admission-plugins=NamespaceLifecycle,LimitRanger,ServiceAccount,DefaultStorageClass,DefaultTolerationSeconds,NodeRestriction,ResourceQuota  \--authorization-mode=Node,RBAC  \--enable-bootstrap-token-auth=true  \--requestheader-client-ca-file=/etc/kubernetes/pki/front-proxy-ca.pem  \--proxy-client-cert-file=/etc/kubernetes/pki/front-proxy-client.pem  \--proxy-client-key-file=/etc/kubernetes/pki/front-proxy-client-key.pem  \--requestheader-allowed-names=aggregator  \--requestheader-group-headers=X-Remote-Group  \--requestheader-extra-headers-prefix=X-Remote-Extra-  \--requestheader-username-headers=X-Remote-User# --token-auth-file=/etc/kubernetes/token.csv
Restart=on-failureRestartSec=10sLimitNOFILE=65535
[Install]WantedBy=multi-user.target

1.3、Master03配置

[Unit]Description=Kubernetes API ServerDocumentation=https://github.com/kubernetes/kubernetesAfter=network.target
[Service]ExecStart=/usr/local/bin/kube-apiserver \--v=2  \--logging-format=text  \--allow-privileged=true  \--bind-address=0.0.0.0  \--secure-port=6443  \--advertise-address=192.168.1.22 \--service-cluster-ip-range=172.16.0.1/12  \--service-node-port-range=10000-40000  \--etcd-servers=https://192.168.1.20:2379,https://192.168.1.21:2379,https://192.168.1.22:2379 \--etcd-cafile=/etc/etcd/ssl/etcd-ca.pem  \--etcd-certfile=/etc/etcd/ssl/etcd.pem  \--etcd-keyfile=/etc/etcd/ssl/etcd-key.pem  \--client-ca-file=/etc/kubernetes/pki/ca.pem  \--tls-cert-file=/etc/kubernetes/pki/apiserver.pem  \--tls-private-key-file=/etc/kubernetes/pki/apiserver-key.pem  \--kubelet-client-certificate=/etc/kubernetes/pki/apiserver.pem  \--kubelet-client-key=/etc/kubernetes/pki/apiserver-key.pem  \--service-account-key-file=/etc/kubernetes/pki/sa.pub  \--service-account-signing-key-file=/etc/kubernetes/pki/sa.key  \--service-account-issuer=https://kubernetes.default.svc.cluster.local \--kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname  \--enable-admission-plugins=NamespaceLifecycle,LimitRanger,ServiceAccount,DefaultStorageClass,DefaultTolerationSeconds,NodeRestriction,ResourceQuota  \--authorization-mode=Node,RBAC  \--enable-bootstrap-token-auth=true  \--requestheader-client-ca-file=/etc/kubernetes/pki/front-proxy-ca.pem  \--proxy-client-cert-file=/etc/kubernetes/pki/front-proxy-client.pem  \--proxy-client-key-file=/etc/kubernetes/pki/front-proxy-client-key.pem  \--requestheader-allowed-names=aggregator  \--requestheader-group-headers=X-Remote-Group  \--requestheader-extra-headers-prefix=X-Remote-Extra-  \--requestheader-username-headers=X-Remote-User# --token-auth-file=/etc/kubernetes/token.csv
Restart=on-failureRestartSec=10sLimitNOFILE=65535
[Install]WantedBy=multi-user.target

1.4、启动apiserver

所有Master节点开启kube-apiserver

systemctl daemon-reload && systemctl enable --now kube-apiserver

检测kube-server状态

systemctl status kube-apiserver

2、ControllerManager

所有Master节点配置kube-controller-manager service（所有master节点配置一样）

注意本文档使用的k8s Pod网段为10.96.0.0/12，该网段不能和宿主机的网段、k8s Service网段的重复，请按需修改

[root@master01 pki]# vim /usr/lib/systemd/system/kube-controller-manager.service

[Unit]Description=Kubernetes Controller ManagerDocumentation=https://github.com/kubernetes/kubernetesAfter=network.target
[Service]ExecStart=/usr/local/bin/kube-controller-manager \--v=2 \--logging-format=text \--bind-address=127.0.0.1 \--root-ca-file=/etc/kubernetes/pki/ca.pem \--cluster-signing-cert-file=/etc/kubernetes/pki/ca.pem \--cluster-signing-key-file=/etc/kubernetes/pki/ca-key.pem \--service-account-private-key-file=/etc/kubernetes/pki/sa.key \--kubeconfig=/etc/kubernetes/controller-manager.kubeconfig \--leader-elect=true \--use-service-account-credentials=true \--node-monitor-grace-period=40s \--node-monitor-period=5s \--node-eviction-rate=0.1 \--controllers=*,bootstrapsigner,tokencleaner \--allocate-node-cidrs=true \--cluster-cidr=10.96.0.0/12 \--requestheader-client-ca-file=/etc/kubernetes/pki/front-proxy-ca.pem \--node-cidr-mask-size=24
Restart=alwaysRestartSec=10s
[Install]WantedBy=multi-user.target

2.1 所有Master节点启动kube-controller-manager

systemctl daemon-reloadsystemctl enable --now kube-controller-manager

3、Scheduler

所有Master节点配置kube-scheduler service（所有master节点配置一样）

# vim /usr/lib/systemd/system/kube-scheduler.service [Unit]Description=Kubernetes SchedulerDocumentation=https://github.com/kubernetes/kubernetesAfter=network.target
[Service]ExecStart=/usr/local/bin/kube-scheduler \--v=2 \--logging-format=text \--bind-address=127.0.0.1 \--leader-elect=true \--kubeconfig=/etc/kubernetes/scheduler.kubeconfig
Restart=alwaysRestartSec=10s
[Install]WantedBy=multi-user.target

systemctl daemon-reloadsystemctl enable --now kube-scheduler

十四、TLS Bootstrapping配置

只需要在Master01创建bootstrap

# 注意，如果不是高可用集群，192.168.1.15:8443改为master01的地址，8443改为apiserver的端口，默认是6443

cd /root/k8s-ha-install/bootstrap
kubectl config set-cluster kubernetes     \  --certificate-authority=/etc/kubernetes/pki/ca.pem     \  --embed-certs=true     \  --server=https://192.168.1.15:8443     \  --kubeconfig=/etc/kubernetes/bootstrap-kubelet.kubeconfig
kubectl config set-credentials tls-bootstrap-token-user     \  --token=c8ad9c.2e4d610cf3e7426e \  --kubeconfig=/etc/kubernetes/bootstrap-kubelet.kubeconfig
kubectl config set-context tls-bootstrap-token-user@kubernetes     \  --cluster=kubernetes     \  --user=tls-bootstrap-token-user     \  --kubeconfig=/etc/kubernetes/bootstrap-kubelet.kubeconfig
kubectl config use-context tls-bootstrap-token-user@kubernetes     \  --kubeconfig=/etc/kubernetes/bootstrap-kubelet.kubeconfig

注意：如果要修改bootstrap.secret.yaml的token-id和token-secret，需要保证下图红圈内的字符串一致的，并且位数是一样的。还要保证上个命令的黄色字体：c8ad9c.2e4d610cf3e7426e与你修改的字符串要一致

1、配置客户端访问

mkdir -p /root/.kube ;cp /etc/kubernetes/admin.kubeconfig /root/.kube/config

2、检查集群状态

可以正常查询集群状态，才可以继续往下，否则不行，需要排查k8s组件是否有故障

[root@master01 pki]# kubectl get csWarning: v1 ComponentStatus is deprecated in v1.19+NAME                 STATUS    MESSAGE   ERRORscheduler            Healthy   ok        controller-manager   Healthy   ok        etcd-0               Healthy   ok

3、配置bootstrap需要的rbac

[root@master01 bootstrap]# kubectl create -f bootstrap.secret.yaml secret/bootstrap-token-c8ad9c createdclusterrolebinding.rbac.authorization.k8s.io/kubelet-bootstrap createdclusterrolebinding.rbac.authorization.k8s.io/node-autoapprove-bootstrap createdclusterrolebinding.rbac.authorization.k8s.io/node-autoapprove-certificate-rotation createdclusterrole.rbac.authorization.k8s.io/system:kube-apiserver-to-kubelet createdclusterrolebinding.rbac.authorization.k8s.io/system:kube-apiserver created

十五、Kubelet配置

1、复制证书到node节点

Master01节点复制证书至Node节点

cd /etc/kubernetes/for NODE in master02 master03 node01 node02;do  ssh $NODE mkdir -p /etc/kubernetes/pki /etc/etcd/ssl /etc/etcd/ssl
for FILE in etcd-ca.pem etcd.pem etcd-key.pem; do  scp /etc/etcd/ssl/$FILE $NODE:/etc/etcd/ssl/donefor FILE in pki/ca.pem pki/ca-key.pem pki/front-proxy-ca.pem bootstrap-kubelet.kubeconfig;do  scp /etc/kubernetes/$FILE $NODE:/etc/kubernetes/${FILE}donedone

2、所有节点创建相关目录

mkdir -p /var/lib/kubelet /var/log/kubernetes /etc/systemd/system/kubelet.service.d /etc/kubernetes/manifests/

3、所有节点配置kubelet service

vim /usr/lib/systemd/system/kubelet.service

[Unit]Description=Kubernetes KubeletDocumentation=https://github.com/kubernetes/kubernetes
[Service]ExecStart=/usr/local/bin/kubelet
Restart=alwaysStartLimitInterval=0RestartSec=10
[Install]WantedBy=multi-user.target

4、所有节点配置kubelet service的配置文件（也可以写到kubelet.service）

vim /etc/systemd/system/kubelet.service.d/10-kubelet.conf

[Service]Environment="KUBELET_KUBECONFIG_ARGS=--bootstrap-kubeconfig=/etc/kubernetes/bootstrap-kubelet.kubeconfig --kubeconfig=/etc/kubernetes/kubelet.kubeconfig"Environment="KUBELET_SYSTEM_ARGS=--cgroup-driver=systemd"Environment="KUBELET_CONFIG_ARGS=--config=/etc/kubernetes/kubelet-conf.yml"Environment="KUBELET_EXTRA_ARGS=--node-labels=node.kubernetes.io/node='' "ExecStart=ExecStart=/usr/local/bin/kubelet $KUBELET_KUBECONFIG_ARGS $KUBELET_CONFIG_ARGS $KUBELET_SYSTEM_ARGS $KUBELET_EXTRA_ARGS

5、创建kubelet的配置文件

注意：如果更改了k8s的service网段，需要更改kubelet-conf.yml 的clusterDNS:配置，改成k8s Service网段的第十个地址，比如172.16.0.10

vim /etc/kubernetes/kubelet-conf.yml

Version: kubelet.config.k8s.io/v1beta1kind: KubeletConfigurationaddress: 0.0.0.0port: 10250readOnlyPort: 10255authentication: anonymous: enabled: false webhook: cacheTTL: 2m0s enabled: true x509: clientCAFile: /etc/kubernetes/pki/ca.pemauthorization: mode: Webhook webhook: cacheAuthorizedTTL: 5m0s cacheUnauthorizedTTL: 30scgroupDriver: systemdcgroupsPerQOS: trueclusterDNS:- 172.16.0.10clusterDomain: cluster.localcontainerLogMaxFiles: 5containerLogMaxSize: 10MicontentType: application/vnd.kubernetes.protobufcpuCFSQuota: truecpuManagerPolicy: nonecpuManagerReconcilePeriod: 10senableControllerAttachDetach: trueenableDebuggingHandlers: trueenforceNodeAllocatable:- podseventBurst: 10eventRecordQPS: 5evictionHard: imagefs.available: 15% memory.available: 100Mi nodefs.available: 10% nodefs.inodesFree: 5%evictionPressureTransitionPeriod: 5m0sfailSwapOn: truefileCheckFrequency: 20shairpinMode: promiscuous-bridgehealthzBindAddress: 127.0.0.1healthzPort: 10248httpCheckFrequency: 20simageGCHighThresholdPercent: 85imageGCLowThresholdPercent: 80imageMinimumGCAge: 2m0siptablesDropBit: 15iptablesMasqueradeBit: 14kubeAPIBurst: 10kubeAPIQPS: 5makeIPTablesUtilChains: truemaxOpenFiles: 1000000maxPods: 110nodeStatusUpdateFrequency: 10soomScoreAdj: -999podPidsLimit: -1registryBurst: 10registryPullQPS: 5resolvConf: /etc/resolv.confrotateCertificates: trueruntimeRequestTimeout: 2m0sserializeImagePulls: truestaticPodPath: /etc/kubernetes/manifestsstreamingConnectionIdleTimeout: 4h0m0ssyncFrequency: 1m0svolumeStatsAggPeriod: 1m0s

6、启动所有节点kubelet

systemctl daemon-reloadsystemctl enable --now kubelet

此时系统日志/var/log/messages显示只有如下信息为正常Unable to update cni config: no networks found in /etc/cni/net.d如果有很多报错日志，或者有大量看不懂的报错，说明kubelet的配置有误，需要检查kubelet配置查看集群状态

7、查看集群状态

@master01 kubernetes]# kubectl get nodeNAME STATUS ROLES AGE VERSIONmaster01 NotReady <none> 31m v1.28.15master02 NotReady <none> 72s v1.28.15master03 NotReady <none> 69s v1.28.15node01 NotReady <none> 72s v1.28.15node02 NotReady <none> 72s v1.28.15

十六、kube-proxy配置

注意，如果不是高可用集群，192.168.1.15:8443改为master01的地址，8443改为apiserver的端口，默认是6443

以下操作只在Master01执行

cd /root/k8s-ha-installkubectl -n kube-system create serviceaccount kube-proxy
kubectl create clusterrolebinding system:kube-proxy  \  --clusterrole system:node-proxier   \  --serviceaccount kube-system:kube-proxy
kubectl apply -f - <<EOFapiVersion: v1kind: Secretmetadata:  name: kube-proxy-token  namespace: kube-system  annotations:    kubernetes.io/service-account.name: kube-proxy  # 关联到 kube-proxy 的 ServiceAccounttype: kubernetes.io/service-account-tokenEOF
JWT_TOKEN=$(kubectl -n kube-system get secret/kube-proxy-token --output=jsonpath='{.data.token}' | base64 -d)
PKI_DIR=/etc/kubernetes/pkiK8S_DIR=/etc/kubernetes
kubectl config set-cluster kubernetes     \  --certificate-authority=/etc/kubernetes/pki/ca.pem     \  --embed-certs=true     \  --server=https://192.168.1.15:8443     \  --kubeconfig=${K8S_DIR}/kube-proxy.kubeconfig
kubectl config set-credentials kubernetes     \  --token=${JWT_TOKEN}     \  --kubeconfig=/etc/kubernetes/kube-proxy.kubeconfig
kubectl config set-context kubernetes    \  --cluster=kubernetes     \  --user=kubernetes     \  --kubeconfig=/etc/kubernetes/kube-proxy.kubeconfig
kubectl config use-context kubernetes     \  --kubeconfig=/etc/kubernetes/kube-proxy.kubeconfig

1、将kubeconfig发送至其他节点

for NODE in master02 master03; doscp /etc/kubernetes/kube-proxy.kubeconfig  $NODE:/etc/kubernetes/kube-proxy.kubeconfigdonefor NODE in node01 node02; doscp /etc/kubernetes/kube-proxy.kubeconfig $NODE:/etc/kubernetes/kube-proxy.kubeconfigdone

2、所有节点添加kube-proxy的配置和service文件

vim /usr/lib/systemd/system/kube-proxy.service

[Unit]Description=Kubernetes Kube ProxyDocumentation=https://github.com/kubernetes/kubernetesAfter=network.target
[Service]ExecStart=/usr/local/bin/kube-proxy \--config=/etc/kubernetes/kube-proxy.yaml \--v=2
Restart=alwaysRestartSec=10s
[Install]WantedBy=multi-user.target

如果更改了集群Pod的网段，需要更改kube-proxy.yaml的clusterCIDR为自己的Pod网段

vim /etc/kubernetes/kube-proxy.yaml

apiVersion: kubeproxy.config.k8s.io/v1alpha1bindAddress: 0.0.0.0clientConnection:  acceptContentTypes: ""  burst: 10  contentType: application/vnd.kubernetes.protobuf  kubeconfig: /etc/kubernetes/kube-proxy.kubeconfig  qps: 5clusterCIDR: 10.96.0.0/12 configSyncPeriod: 15m0sconntrack:  max: null  maxPerCore: 32768  min: 131072  tcpCloseWaitTimeout: 1h0m0s  tcpEstablishedTimeout: 24h0m0senableProfiling: falsehealthzBindAddress: 0.0.0.0:10256hostnameOverride: ""iptables:  masqueradeAll: false  masqueradeBit: 14  minSyncPeriod: 0s  syncPeriod: 30sipvs:  masqueradeAll: true  minSyncPeriod: 5s  scheduler: "rr"  syncPeriod: 30skind: KubeProxyConfigurationmetricsBindAddress: 127.0.0.1:10249mode: "ipvs"nodePortAddresses: nulloomScoreAdj: -999portRange: ""udpIdleTimeout: 250ms

3、所有节点启动kube-proxy

 systemctl daemon-reload systemctl enable --now kube-proxy

十七、安装Calico

1、以下步骤只在master01执行

cd /root/k8s-ha-install/calico/
# 更改calico的网段，主要需要将红色部分的网段，改为自己的Pod网段sed -i "s#POD_CIDR#10.96.0.0/12#g" calico.yaml

2、检查网段是自己的Pod网段

cd /root/k8s-ha-install/calico/
# 更改calico的网段，主要需要将红色部分的网段，改为自己的Pod网段sed -i "s#POD_CIDR#10.96.0.0/12#g" calico.yaml

3、创建

kubectl apply -f calico.yaml

4、查看容器状态

[root@master01 CoreDNS]# kubectl get po -nkube-systemNAME                                       READY   STATUS    RESTARTS   AGEcalico-kube-controllers-6d48795585-gclj7   1/1     Running   0          5m54scalico-node-482wx                          1/1     Running   0          5m54scalico-node-j7p69                          1/1     Running   0          5m54scalico-node-mjr89                          1/1     Running   0          5m54scalico-node-w8v4q                          1/1     Running   0          5m54scalico-node-wm4nc                          1/1     Running   0          5m54s

5、查看集群是否ready

[root@master01 CoreDNS]# kubectl get nodeNAME       STATUS   ROLES    AGE   VERSIONmaster01   Ready    <none>   22h   v1.28.15master02   Ready    <none>   21h   v1.28.15master03   Ready    <none>   21h   v1.28.15node01     Ready    <none>   21h   v1.28.15node02     Ready    <none>   21h   v1.28.15

十八、安装CoreDNS

如果更改了k8s service的网段需要将coredns的serviceIP改成k8s service网段的第十个IP

cd /root/k8s-ha-install/
COREDNS_SERVICE_IP=`kubectl get svc | grep kubernetes | awk '{print $3}'`0sed -i "s#KUBEDNS_SERVICE_IP#${COREDNS_SERVICE_IP}#g" CoreDNS/coredns.yaml
kubectl  create -f CoreDNS/coredns.yaml

[root@master01 CoreDNS]# kubectl get po -nkube-systemNAME                                       READY   STATUS    RESTARTS   AGEcoredns-788958459b-ncr8x                   1/1     Running   0          11s

十九、安装Metrics Server

在新版的Kubernetes中系统资源的采集均使用Metrics-server，可以通过Metrics采集节点和Pod的内存、磁盘、CPU和网络的使用率。

cd /root/k8s-ha-install/metrics-serverkubectl  create -f .

等待metrics server启动然后查看状态

[root@master01 dashboard]# kubectl top nodeNAME       CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%   master01   309m         3%     2409Mi          30%       master02   218m         2%     2040Mi          25%       master03   274m         3%     2059Mi          26%       node01     97m          1%     1137Mi          14%       node02     103m         1%     1229Mi          15%

二十、安装dashboard

1、安装

Dashboard用于展示集群中的各类资源，同时也可以通过Dashboard实时查看Pod的日志和在容器中执行一些命令等。

cd /root/k8s-ha-install/dashboard/kubectl  create -f .

2、登录dashboard

[root@master01 dashboard]# kubectl get svc -nkubernetes-dashboardNAME                        TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)         AGEdashboard-metrics-scraper   ClusterIP   172.19.236.49   <none>        8000/TCP        28mkubernetes-dashboard        NodePort    172.17.107.71   <none>        443:20110/TCP   28m

3、查看token值

kubectl -n kube-system describe secret $(kubectl -n kube-system get secret | grep admin-user | awk '{print $1}')

4、登录效果

查看全文

http://www.dtcms.com/a/186212.html

Windows 添加 hosts 映射

ES6基础特性

关于高并发GIS数据处理的一点经验分享

apt 软件源与 Docker 镜像源

【Linux网络】 HTTP cookie与session

【中药快速划价小程序】佳易王中药方快速划价软件：主打一个快字，中药划价软件操作指南，软件实操教程

交易所开发-如何开发一个交易所

AbMole Olaparib：打破常规，用PARP抑制重塑肿瘤研究

日语学习-日语知识点小记-构建基础-JLPT-N4阶段（1９）：て形/ないで

MySQL的sql_mode详解：从优雅草分发平台故障谈数据库模式配置-优雅草卓伊凡

【数据结构入门训练DAY-32】LETTERS

MySQL 8.0 OCP（1Z0-908）英文题库(21-30)

Pytorch常用统计和矩阵运算

c/c++爬虫总结

E9 泛微OA获取requestid

【node】6 包与npm

MINIX 1.0 文件系统的实现(C/C++实现)

Kotlin跨平台Compose Multiplatform实战指南

传导发射中的模拟手

BGP练习

openwrt目录结构（部分）

深入理解目标检测中的关键指标及其计算方法

Mosaic数据增强技术

构造+简单树状

Java 并发编程挑战：从原理到实战的深度剖析与解决方案

pnpm使用报错

https的发展历程

SpringBoot医院病房信息管理系统开发实现

C++：公有，保护及私有继承

字节开源FlowGram与n8n 技术选型

一、节点规划

二、版本选择

三、hosts配置

1、master01配置

2、同步至其他主机

四、SSH配置

1、所有节点初始化SSH配置

2、master01免密登录其他节点

五、修改yum源为aliyun

六、CentOS基本环境升级

1、安装基础工具

2、所有节点关闭firewalld 、dnsmasq、selinux

3、所有节点关闭swap分区，fstab注释swap

4、所有节点同步时间

4.1、安装ntpdate

4.2、同步时间

5、所有节点配置limit

5.1、ulimit -SHn 65535

5.2、limits.conf

6、升级内核

6.1、下载

6.2、同步至其他节点​​​​​​

6.3、所有节点安装

6.4、更改内核启动顺序

6.5、检查默认内核是不是4.19

6.6、所有节点重启，然后检查内核是不是4.19

6.7、所有节点安装ipvsadm

6.7.1、所有节点配置ipvs模块

6.7.2、ipvs.conf

6.7.3、检查是否加载

7、内核参数配置

8、所有节点配置完后，重启节点，保证服务重启后，配置依旧

七、基本组件安装

1、Containerd作为Runtime

2、所有节点配置Containerd需要的模块

3、所有节点配置Containerd所需要的内核

4、所有节点配置Containerd的配置文件

5、所有节点将Containerd的Cgroup改为Systemd

6、所有节点将sandbox_image的Pause镜像

7、所有节点启动Containerd，并配置开机自启动

8、所有节点配置crictl客户端连接的运行时位置

八、ETCD安装

1、下载

2、解压ETCD安装包，安装至执行目录

3、同步执行文件至master02、master03节点

九、K8S安装

1、下载

2、解压安装包至执行目录

3、同步执行文件到其他节点

十、生成证书

1、下载生成证书工具

2、ETCD证书

2.1、所有Master节点创建etcd证书目录

2.2、clone相关配置文件

2.3、生成ETCD CA证书和CA证书的key

2.4、复制证书至其他master节点

3、K8S组件证书

3.1、Master01生成kubernetes证书

3.2、生成apiserver的聚合证书

3.3、生成controller-manage的证书

3.4、set-cluster：设置一个集群项

3.5、设置一个环境项，一个上下文

3.6、set-credentials 设置一个用户项

3.7、使用某个环境当做默认环境

3.8、生成kube-scheduler证书

3.9、配置admin证书

3.10、创建ServiceAccount Key -> secret

4、将证书发送至其他节点

十一、ETCD配置

1、master01配置

2、master02配置

3、master03配置

4、创建service

5、查看ETCD的状态

十二、高可用配置

1、所有Master节点安装keepalived和haproxy

2、所有Master配置HAProxy，配置一样

3、Master01 keepalived

4、Master02 keepalived

5、Master03 keepalived

6.2、同步至其他节点

1、以下步骤只在master01执行

2、检查网段是自己的Pod网段