init commit for dev-3.0

.gitignore

@@ -2,12 +2,8 @@ down/*
 !down/download.sh
 !down/offline_images
 
-# k8s binaries directory
+# binaries directory
 bin/*
-!bin/readme.md
-
-# ansible hosts
-hosts
 
 # k8s storage manifests 
 manifests/storage/*
@@ -18,8 +14,8 @@ roles/cluster-backup/files/*
 !roles/cluster-backup/files/readme.md
 
 # role based variable settings, exclude roles/os-harden/vars/
-/roles/*/vars/*
+#/roles/*/vars/*
-!/roles/os-harden/vars/
+#!/roles/os-harden/vars/
 
 # cluster backups
 .cluster/

ansible.cfg

@@ -32,7 +32,7 @@
 gathering = smart 
 
 # additional paths to search for roles in, colon separated
-roles_path    = /etc/ansible/roles
+roles_path    = /etc/kubeasz/roles
 
 # uncomment this to disable SSH key host checking
 host_key_checking = False

dockerfiles/readme.md

@@ -1,3 +0,0 @@
-# Dockerfiles for building images needed
-
-Please refer to https://github.com/kubeasz/dockerfiles

down/download.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 # This script describes where to download the official released binaries needed
-# It's suggested to download using 'tools/easzup -D', everything needed will be ready in '/etc/ansible'
+# It's suggested to download using 'ezdown -D', everything needed will be ready in '/etc/kubeasz'
 
 # example releases
 K8S_VER=v1.13.7
@@ -11,7 +11,7 @@ DOCKER_COMPOSE_VER=1.23.2
 HARBOR_VER=v1.9.4
 CONTAINERD_VER=1.2.6
 
-echo -e "\nNote: It's strongly recommended that downloading with 'tools/easzup -D', everything needed will be ready in '/etc/ansible'."
+echo -e "\nNote: It's strongly recommended that downloading with 'ezdown -D', everything needed will be ready in '/etc/kubeasz'."
 
 echo -e "\n----download k8s binary at:"
 echo -e https://dl.k8s.io/${K8S_VER}/kubernetes-server-linux-amd64.tar.gz

example/config.yml

@@ -0,0 +1,182 @@
+############################
+# role:prepare
+############################
+# 可选离线安装系统软件包 (offline|online)
+INSTALL_SOURCE: "online"
+
+
+############################
+# role:chrony
+############################
+# 设置时间源服务器
+ntp_servers:
+  - "ntp1.aliyun.com"
+  - "time1.cloud.tencent.com"
+  - "0.cn.pool.ntp.org"
+
+# 设置允许内部时间同步的网络段，比如"10.0.0.0/8"，默认全部允许
+local_network: "0.0.0.0/0"
+
+
+############################
+# role:deploy
+############################
+# default: ca will expire in 100 years
+# default: certs issued by the ca will expire in 50 years
+CA_EXPIRY: "876000h"
+CERT_EXPIRY: "438000h"
+
+# kubeconfig 配置参数，注意权限根据‘USER_NAME’设置：
+# 'admin' 表示创建集群管理员（所有）权限的 kubeconfig
+# 'read' 表示创建只读权限的 kubeconfig
+CLUSTER_NAME: "cluster1"
+USER_NAME: "admin"
+CONTEXT_NAME: "context-{{ CLUSTER_NAME }}-{{ USER_NAME }}"
+
+
+############################
+# role:runtime [containerd,docker]
+############################
+# [.]启用容器仓库镜像
+ENABLE_MIRROR_REGISTRY: true
+
+# [containerd]基础容器镜像
+SANDBOX_IMAGE: "easzlab/pause-amd64:3.2"
+
+# [containerd]容器持久化存储目录
+CONTAINERD_STORAGE_DIR: "/var/lib/containerd"
+
+# [docker]容器存储目录
+DOCKER_STORAGE_DIR: "/var/lib/docker"
+
+# [docker]开启Restful API
+ENABLE_REMOTE_API: false
+
+# [docker]信任的HTTP仓库
+INSECURE_REG: '["127.0.0.1/8"]'
+
+
+############################
+# role:kube-master
+############################
+# k8s 集群 master 节点证书配置，可以添加多个ip和域名（比如增加公网ip和域名）
+MASTER_CERT_HOSTS:
+  - "10.1.1.1"
+  - "k8s.test.io"
+  #- "www.test.com"
+
+# node 节点上 pod 网段掩码长度（决定每个节点最多能分配的pod ip地址）
+# 如果flannel 使用 --kube-subnet-mgr 参数，那么它将读取该设置为每个节点分配pod网段
+# https://github.com/coreos/flannel/issues/847
+NODE_CIDR_LEN: 24
+
+
+############################
+# role:kube-node
+############################
+# Kubelet 根目录
+KUBELET_ROOT_DIR: "/var/lib/kubelet"
+
+# node节点最大pod 数
+MAX_PODS: 110
+
+# 配置为kube组件（kubelet,kube-proxy,dockerd等）预留的资源量
+# 数值设置详见templates/kubelet-config.yaml.j2
+KUBE_RESERVED_ENABLED: "yes"
+
+# k8s 官方不建议草率开启 system-reserved, 除非你基于长期监控，了解系统的资源占用状况；
+# 并且随着系统运行时间，需要适当增加资源预留，数值设置详见templates/kubelet-config.yaml.j2
+# 系统预留设置基于 4c/8g 虚机，最小化安装系统服务，如果使用高性能物理机可以适当增加预留
+# 另外，集群安装时候apiserver等资源占用会短时较大，建议至少预留1g内存
+SYS_RESERVED_ENABLED: "no"
+
+# haproxy balance mode
+BALANCE_ALG: "roundrobin"
+
+
+############################
+# role:network [flannel,calico,cilium,kube-ovn,kube-router]
+############################
+# [flannel]设置flannel 后端"host-gw","vxlan"等
+FLANNEL_BACKEND: "vxlan"
+DIRECT_ROUTING: false
+
+# [flannel] flanneld_image: "quay.io/coreos/flannel:v0.10.0-amd64"
+flannelVer: "v0.13.0-amd64"
+flanneld_image: "easzlab/flannel:{{ flannelVer }}"
+
+# [flannel]离线镜像tar包
+flannel_offline: "flannel_{{ flannelVer }}.tar"
+
+# [calico]设置 CALICO_IPV4POOL_IPIP=“off”,可以提高网络性能，条件限制详见 docs/setup/calico.md
+CALICO_IPV4POOL_IPIP: "Always"
+
+# [calico]设置 calico-node使用的host IP，bgp邻居通过该地址建立，可手工指定也可以自动发现
+IP_AUTODETECTION_METHOD: "can-reach={{ groups['kube-master'][0] }}"
+
+# [calico]设置calico 网络 backend: brid, vxlan, none
+CALICO_NETWORKING_BACKEND: "brid"
+
+# [calico]更新支持calico 版本: [v3.3.x] [v3.4.x] [v3.8.x] [v3.15.x]
+calicoVer: "v3.15.3"
+calico_ver: "{{ calicoVer }}"
+
+# [calico]calico 主版本
+calico_ver_main: "{{ calico_ver.split('.')[0] }}.{{ calico_ver.split('.')[1] }}"
+
+# [calico]离线镜像tar包
+calico_offline: "calico_{{ calico_ver }}.tar"
+
+# [cilium]CILIUM_ETCD_OPERATOR 创建的 etcd 集群节点数 1,3,5,7...
+ETCD_CLUSTER_SIZE: 1
+
+# [cilium]镜像版本
+cilium_ver: "v1.4.1"
+
+# [cilium]离线镜像tar包
+cilium_offline: "cilium_{{ cilium_ver }}.tar"
+
+# [kube-ovn]选择 OVN DB and OVN Control Plane 节点，默认为第一个master节点
+OVN_DB_NODE: "{{ groups['kube-master'][0] }}"
+
+# [kube-ovn]离线镜像tar包
+kube_ovn_offline: "kube_ovn_0.9.1.tar"
+
+# [kube-router]公有云上存在限制，一般需要始终开启 ipinip；自有环境可以设置为 "subnet"
+OVERLAY_TYPE: "full"
+
+# [kube-router]NetworkPolicy 支持开关
+FIREWALL_ENABLE: "true"
+
+# [kube-router]kube-router 镜像版本
+kube_router_ver: "v0.3.1"
+busybox_ver: "1.28.4"
+
+# [kube-router]kube-router 离线镜像tar包
+kuberouter_offline: "kube-router_{{ kube_router_ver }}.tar"
+busybox_offline: "busybox_{{ busybox_ver }}.tar"
+
+
+############################
+# role:cluster-addon
+############################
+# dns 自动安装，'dns_backend'可选"coredns"和“kubedns”
+dns_install: "yes"
+dns_backend: "coredns"
+corednsVer: "1.7.1"
+
+# metric server 自动安装
+metricsserver_install: "yes"
+metricsVer: "v0.3.6"
+
+# dashboard 自动安装
+dashboard_install: "yes"
+dashboardVer: "v2.1.0"
+dashboardMetricsScraperVer: "v1.0.6"
+
+
+############################
+# role:harbor
+############################
+# harbor version，完整版本号
+HARBOR_VER: "v1.9.4"

example/hosts.allinone

@@ -53,8 +53,11 @@ CLUSTER_DNS_DOMAIN="cluster.local."
 # Binaries Directory
 bin_dir="/opt/kube/bin"
 
-# CA and other components cert/key Directory
-ca_dir="/etc/kubernetes/ssl"
-
 # Deploy Directory (kubeasz workspace)
-base_dir="/etc/ansible"
+base_dir="/etc/kubeasz"
+
+# Directory for a specific cluster
+cluster_dir="{{ base_dir }}/clusters/_cluster_name_"
+
+# CA and other components cert/key Directory
+ca_dir="{{ cluster_dir }}/ssl"

example/hosts.multi-node

@@ -61,4 +61,7 @@ bin_dir="/opt/kube/bin"
 ca_dir="/etc/kubernetes/ssl"
 
 # Deploy Directory (kubeasz workspace)
-base_dir="/etc/ansible"
+base_dir="/etc/kubeasz"
+
+# Directory for a specific cluster
+cluster_dir="{{ base_dir }}/clusters/_cluster_name_"

ezctl

@@ -1,38 +1,54 @@
 #!/bin/bash
-# 
+#  Create & manage k8s clusters by 'kubeasz'
-# This script aims to manage k8s clusters created by 'kubeasz'. (developing)
 
 set -o nounset
 set -o errexit
 #set -o xtrace
 
 function usage() {
+    echo -e "\033[33mUsage:\033[0m ezctl COMMAND [args]"
     cat <<EOF
-Usage: easzctl COMMAND [args]
+-------------------------------------------------------------------------------------
+Cluster setups:
+    list		             to list all of the managed clusters
+    new         <cluster>            to start a new k8s deploy with name 'cluster'
+    setup       <cluster>  [step]    to setup a cluster, also supporting a step-by-step way
+    start-aio		             to quickly setup an all-in-one cluster with 'default' settings
 
-Cluster-wide operation:
+Cluster ops:
-    checkout		To switch to context <clustername>, or create it if not existed
+    add-etcd    <cluster>  <args>    to add a etcd-node to the etcd cluster
-    destroy		To destroy the current cluster, '--purge' to also delete the context
+    add-master  <cluster>  <args>    to add a master node to the k8s cluster
-    list		To list all of clusters managed
+    add-node    <cluster>  <args>    to add a work node to the k8s cluster
-    setup		To setup a cluster using the current context
+    del-etcd    <cluster>  <ip>      to delete a etcd-node from the etcd cluster
-    start-aio		To quickly setup an all-in-one cluster for testing (like minikube)
+    del-master  <cluster>  <ip>      to delete a master node from the k8s cluster
+    del-node    <cluster>  <ip>      to delete a work node from the k8s cluster
+    upgrade     <cluster>            to upgrade the k8s cluster
+    destroy     <cluster>            to destroy the current cluster, '--purge' to also delete the context
 
-In-cluster operation:
+Use "ezctl help <command>" for more information about a given command.
-    add-etcd		To add a etcd-node to the etcd cluster
-    add-master		To add a kube-master(master node) to the k8s cluster
-    add-node		To add a kube-node(work node) to the k8s cluster
-    del-etcd		To delete a etcd-node from the etcd cluster
-    del-master		To delete a kube-master from the k8s cluster
-    del-node		To delete a kube-node from the k8s cluster
-    upgrade		To upgrade the k8s cluster
-
-Extra operation:
-    basic-auth   	To enable/disable basic-auth for apiserver
-
-Use "easzctl help <command>" for more information about a given command.
 EOF
 }
 
+function logger() {
+  TIMESTAMP=$(date +'%Y-%m-%d %H:%M:%S')
+  case "$1" in
+    debug)
+      echo -e "$TIMESTAMP \033[36mDEBUG\033[0m $2"
+      ;;
+    info)
+      echo -e "$TIMESTAMP \033[32mINFO\033[0m $2"
+      ;;
+    warn)
+      echo -e "$TIMESTAMP \033[33mWARN\033[0m $2"
+      ;;
+    error)
+      echo -e "$TIMESTAMP \033[31mERROR\033[0m $2"
+      ;;
+    *)
+      ;;
+  esac
+}
+
 function help-info() {
     case "$1" in
         (add-etcd)
@@ -53,9 +69,6 @@ function help-info() {
         (del-node)
             echo -e "Usage: easzctl del-node <node_ip>\n\nread 'https://github.com/easzlab/kubeasz/blob/master/docs/op/op-node.md'"
             ;;
-        (basic-auth)
-            echo -e "Usage: easzctl basic-auth <options>\nOption:\t -s enable basic-auth\n\t -S disable basic-auth\n\t -u <user> set username\n\t -p <pass> set password"
-            ;;
         (*)
             usage
             return 0
@@ -63,12 +76,59 @@ function help-info() {
     esac
 }
 
-function process_cmd() {
+### Cluster setups functions ##############################
-    echo -e "[INFO] \033[33m$ACTION\033[0m : $CMD"
+
-    $CMD || { echo -e "[ERROR] \033[31mAction failed\033[0m : $CMD"; return 1; }
+function new() {
-    echo -e "[INFO] \033[32mAction successed\033[0m : $CMD"
+    # check if already existed
+    [[ -d "clusters/$1" ]] && { logger error "cluster: $1 already existed"; exit 1; }
+    [[ "$1" == default ]] && { logger error "name 'default' is reserved for `ezctl start-aio`"; exit 1; }
+
+    logger debug "generate custom cluster files in clusters/$1"
+    mkdir -p "clusters/$1"
+    cp example/hosts.multi-node "clusters/$1/hosts"
+    sed -i "s/_cluster_name_/$1/g" "clusters/$1/hosts"
+    cp example/config.yml "clusters/$1/config.yml"
+
+    logger debug "cluster $1: files successfully created."
+    logger info "next steps 1: to config 'clusters/$1/hosts'"
+    logger info "next steps 2: to config 'clusters/$1/config.yml'"
 }
 
+function setup() {
+    [[ -d "clusters/$1" ]] || { logger error "invalid config, run 'ezctl new $1' first"; return 1; }
+    [[ -f "bin/kube-apiserver" ]] || { logger error "no binaries founded, run 'ezdown -D' fist"; return 1; }
+
+    logger info "\n cluster:$1 setup begins in 5s, press any key to abort\n:"
+    ! (read -t5 -n1 ANS) || { logger warn "setup aborted"; return 1; }
+
+    ansible-playbook -i "clusters/$1/hosts" -e "@clusters/$1/config.yml" playbooks/90.setup.yml || return 1
+}
+
+function list() {
+    [ -f "$BASEPATH/.cluster/current_cluster" ] || { echo "[ERROR] invalid context, run 'easzctl checkout <cluster_name>' first"; return 1; }
+    CLUSTER=$(cat $BASEPATH/.cluster/current_cluster)
+    echo -e "\nlist of managed contexts (current: \033[33m$CLUSTER\033[0m)"
+    i=1; for c in $(ls $BASEPATH/.cluster/ |grep -Ev "backup|ssl|current_cluster|kubeconfig");
+    do
+        echo -e "==> context $i:\t$c"
+        let "i++"
+    done
+    echo -e "\nlist of installed clusters (current: \033[33m$CLUSTER\033[0m)"
+    i=1; for c in $(ls $BASEPATH/.cluster/ |grep -Ev "backup|ssl|current_cluster|kubeconfig");
+    do
+        KUBECONF=$BASEPATH/.cluster/$c/config
+        if [ -f "$KUBECONF" ]; then
+            echo -e "==> cluster $i:\t$c"
+            $BASEPATH/bin/kubectl --kubeconfig=$KUBECONF get node
+        fi
+        let "i++"
+    done
+}
+
+
+
+
+
 ### in-cluster operation functions ##############################
 
 function add-node() {
@@ -256,34 +316,6 @@ function install_context() {
     fi
 }
 
-function checkout() {
-    # check directory '.cluster', initialize it if not existed
-    if [ ! -f "$BASEPATH/.cluster/current_cluster" ]; then
-        echo "[INFO] initialize directory $BASEPATH/.cluster"
-        mkdir -p $BASEPATH/.cluster/default
-        echo default > $BASEPATH/.cluster/current_cluster
-    fi
-
-    # check if $1 is already the current context
-    CLUSTER=$(cat $BASEPATH/.cluster/current_cluster)
-    [ "$1" != "$CLUSTER" ] || { echo "[WARN] $1 is already the current context"; return 0; }
-
-    echo "[INFO] save current context: $CLUSTER"
-    save_context
-    echo "[INFO] clean context: $CLUSTER"
-    rm -rf $BASEPATH/hosts /root/.kube/* $BASEPATH/.cluster/ssl $BASEPATH/.cluster/kube-proxy.kubeconfig
-
-    # check context $1, install it if existed, otherwise initialize it using default context
-    if [ ! -d "$BASEPATH/.cluster/$1" ];then
-        echo "[INFO] context $1 not existed, initialize it using default context"
-        cp -rp $BASEPATH/.cluster/default $BASEPATH/.cluster/$1
-        rm -f $BASEPATH/.cluster/$1/hosts $BASEPATH/.cluster/$1/config
-    fi
-    echo "[INFO] change current context to $1"
-    echo $1 > $BASEPATH/.cluster/current_cluster
-    install_context;
-}
-
 function setup() {
     [ -f "$BASEPATH/.cluster/current_cluster" ] || { echo "[ERROR] invalid context, run 'easzctl checkout <cluster_name>' first"; return 1; }
     [ -f "$BASEPATH/bin/kube-apiserver" ] || { echo "[ERROR] no binaries found, download then fist"; return 1; }
@@ -405,7 +437,7 @@ function basic-auth(){
 
 ### Main Lines ##################################################
 
-BASEPATH=/etc/ansible
+BASEPATH=/etc/kubeasz
 
 [ "$#" -gt 0 ] || { usage >&2; exit 2; }
 

ezdown

@@ -2,9 +2,9 @@
 #--------------------------------------------------
 # This script is for: 
 # 1. to download the scripts/binaries/images needed for installing a k8s cluster with kubeasz
-# 2. to run kubeasz in a container (optional, not recommend)
+# 2. to run kubeasz in a container (optional)
 # @author:   gjmzj
-# @usage:    ./easzup
+# @usage:    ./ezdown
 # @repo:     https://github.com/easzlab/kubeasz
 # @ref:      https://github.com/kubeasz/dockerfiles
 #--------------------------------------------------
@@ -12,7 +12,7 @@ set -o nounset
 set -o errexit
 #set -o xtrace
 
-# default version, can be overridden by cmd line options, see ./easzup
+# default version, can be overridden by cmd line options, see usage
 DOCKER_VER=19.03.14
 KUBEASZ_VER=2.2.3
 K8S_BIN_VER=v1.20.1
@@ -28,38 +28,75 @@ flannelVer=v0.13.0-amd64
 metricsVer=v0.3.6
 pauseVer=3.2
 
+function usage() {
+  echo -e "\033[33mUsage:\033[0m ezdown [options] [args]"
+  cat <<EOF
+  option: -{DdekSz}
+    -C         stop&clean all local containers
+    -D         download all into "$BASE"
+    -P         download system packages for offline installing
+    -S         start kubeasz in a container
+    -d <ver>   set docker-ce version, default "$DOCKER_VER"
+    -e <ver>   set kubeasz-ext-bin version, default "$EXT_BIN_VER"
+    -k <ver>   set kubeasz-k8s-bin version, default "$K8S_BIN_VER"
+    -m <str>   set docker registry mirrors, default "CN"(used in Mainland,China)
+    -p <ver>   set kubeasz-sys-pkg version, default "$SYS_PKG_VER"
+    -z <ver>   set kubeasz version, default "$KUBEASZ_VER"
+
+see more at https://github.com/kubeasz/dockerfiles
+EOF
+}
+
+function logger() {
+  TIMESTAMP=$(date +'%Y-%m-%d %H:%M:%S')
+  case "$1" in
+    debug)
+      echo -e "$TIMESTAMP \033[36mDEBUG\033[0m $2"
+      ;;
+    info)
+      echo -e "$TIMESTAMP \033[32mINFO\033[0m $2"
+      ;;
+    warn)
+      echo -e "$TIMESTAMP \033[33mWARN\033[0m $2"
+      ;;
+    error)
+      echo -e "$TIMESTAMP \033[31mERROR\033[0m $2"
+      ;;
+    *)
+      ;;
+  esac
+}
+
 function download_docker() {
-  echo -e "[INFO] \033[33mdownloading docker binaries\033[0m $DOCKER_VER"
   if [[ "$REGISTRY_MIRROR" == CN ]];then
     DOCKER_URL="https://mirrors.tuna.tsinghua.edu.cn/docker-ce/linux/static/stable/x86_64/docker-${DOCKER_VER}.tgz"
   else
     DOCKER_URL="https://download.docker.com/linux/static/stable/x86_64/docker-${DOCKER_VER}.tgz"
   fi
 
-  mkdir -p /opt/kube/bin /etc/ansible/down /etc/ansible/bin
+  if [[ -f "$BASE/down/docker-${DOCKER_VER}.tgz" ]];then
-  if [[ -f "/etc/ansible/down/docker-${DOCKER_VER}.tgz" ]];then
+    logger warn "docker binaries already existed"
-    echo "[INFO] docker binaries already existed"
   else
-    echo -e "[INFO] \033[33mdownloading docker binaries\033[0m $DOCKER_VER"
+    logger info "downloading docker binaries, version $DOCKER_VER"
     if [[ -e /usr/bin/curl ]];then
-      curl -C- -O --retry 3 "$DOCKER_URL" || { echo "[ERROR] downloading docker failed"; exit 1; }
+      curl -C- -O --retry 3 "$DOCKER_URL" || { logger error "downloading docker failed"; exit 1; }
     else
-      wget -c "$DOCKER_URL" || { echo "[ERROR] downloading docker failed"; exit 1; }
+      wget -c "$DOCKER_URL" || { logger error "downloading docker failed"; exit 1; }
     fi
-    /bin/mv -f ./docker-${DOCKER_VER}.tgz /etc/ansible/down
+    /bin/mv -f ./docker-${DOCKER_VER}.tgz $BASE/down
   fi
 
-  tar zxf /etc/ansible/down/docker-${DOCKER_VER}.tgz -C /etc/ansible/down && \
+  tar zxf $BASE/down/docker-${DOCKER_VER}.tgz -C $BASE/down && \
-  /bin/cp -f /etc/ansible/down/docker/* /etc/ansible/bin && \
+  /bin/cp -f $BASE/down/docker/* $BASE/bin && \
-  /bin/mv -f /etc/ansible/down/docker/* /opt/kube/bin && \
+  /bin/mv -f $BASE/down/docker/* /opt/kube/bin && \
   ln -sf /opt/kube/bin/docker /bin/docker 
 }
 
 function install_docker() {
   # check if a container runtime is already installed
-  systemctl status docker|grep Active|grep -q running && { echo "[WARN] docker is already running."; return 0; }
+  systemctl status docker|grep Active|grep -q running && { logger warn "docker is already running."; return 0; }
  
-  echo "[INFO] generate docker service file"
+  logger debug "generate docker service file"
   cat > /etc/systemd/system/docker.service << EOF
 [Unit]
 Description=Docker Application Container Engine
@@ -82,9 +119,9 @@ EOF
 
   # configuration for dockerd
   mkdir -p /etc/docker
-  echo "[INFO] generate docker config file"
+  logger debug "generate docker config: /etc/docker/daemon.json"
   if [[ "$REGISTRY_MIRROR" == CN ]];then
-    echo "[INFO] prepare register mirror for $REGISTRY_MIRROR"
+    logger debug "prepare register mirror for $REGISTRY_MIRROR"
     cat > /etc/docker/daemon.json << EOF
 {
   "registry-mirrors": [
@@ -102,7 +139,7 @@ EOF
 }
 EOF
   else
-    echo "[INFO] standard config without registry mirrors"
+    logger debug "standard config without registry mirrors"
     cat > /etc/docker/daemon.json << EOF
 {
   "max-concurrent-downloads": 10,
@@ -118,87 +155,88 @@ EOF
   fi
 
   if [[ -e /etc/centos-release || -e /etc/redhat-release ]]; then
-    echo "[INFO] turn off selinux in CentOS/Redhat"
+    logger debug "turn off selinux in CentOS/Redhat"
     getenforce|grep Disabled || setenforce 0
     sed -i 's/^SELINUX=.*$/SELINUX=disabled/g' /etc/selinux/config
   fi
 
-  echo "[INFO] enable and start docker"
+  logger debug "enable and start docker"
   systemctl enable docker
   systemctl daemon-reload && systemctl restart docker && sleep 4
 }
 
 function get_kubeasz() {
   # check if kubeasz already existed
-  [[ -d "/etc/ansible/roles/kube-node" ]] && { echo "[WARN] kubeasz already existed"; return 0; }
+  [[ -d "$BASE/roles/kube-node" ]] && { logger warn "kubeasz already existed"; return 0; }
 
-  echo -e "[INFO] \033[33mdownloading kubeasz\033[0m $KUBEASZ_VER"
+  logger info "downloading kubeasz: $KUBEASZ_VER"
-  echo "[INFO] run a temporary container"
+  logger debug " run a temporary container"
-  docker run -d --name temp_easz easzlab/kubeasz:${KUBEASZ_VER} || { echo "[ERROR] download failed."; exit 1; }
+  docker run -d --name temp_easz easzlab/kubeasz:${KUBEASZ_VER} || { logger error "download failed."; exit 1; }
 
-  [[ -f "/etc/ansible/down/docker-${DOCKER_VER}.tgz" ]] && /bin/mv -f "/etc/ansible/down/docker-${DOCKER_VER}.tgz" /tmp
+  [[ -f "$BASE/down/docker-${DOCKER_VER}.tgz" ]] && /bin/mv -f "$BASE/down/docker-${DOCKER_VER}.tgz" /tmp
-  [[ -d "/etc/ansible/bin" ]] && /bin/mv -f /etc/ansible/bin /tmp
+  [[ -d "$BASE/bin" ]] && /bin/mv -f $BASE/bin /tmp
 
-  rm -rf /etc/ansible && \
+  rm -rf $BASE && \
-  echo "[INFO] cp kubeasz code from the temporary container" && \
+  logger debug "cp kubeasz code from the temporary container" && \
-  docker cp temp_easz:/etc/ansible /etc/ansible && \
+  docker cp temp_easz:$BASE $BASE && \
-  echo "[INFO] stop&remove temporary container" && \
+  logger debug "stop&remove temporary container" && \
   docker rm -f temp_easz
 
- [[ -f "/tmp/docker-${DOCKER_VER}.tgz" ]] && /bin/mv -f "/tmp/docker-${DOCKER_VER}.tgz" /etc/ansible/down
+  mkdir -p $BASE/bin
- [[ -d "/tmp/bin" ]] && /bin/mv -f /tmp/bin/* /etc/ansible/bin
+  [[ -f "/tmp/docker-${DOCKER_VER}.tgz" ]] && /bin/mv -f "/tmp/docker-${DOCKER_VER}.tgz" $BASE/down
+  [[ -d "/tmp/bin" ]] && /bin/mv -f /tmp/bin/* $BASE/bin
   return 0
 }
 
 function get_k8s_bin() {
-  [[ -f "/etc/ansible/bin/kubelet" ]] && { echo "[WARN] kubernetes binaries existed"; return 0; }
+  [[ -f "$BASE/bin/kubelet" ]] && { logger warn "kubernetes binaries existed"; return 0; }
   
-  echo -e "[INFO] \033[33mdownloading kubernetes\033[0m $K8S_BIN_VER binaries"
+  logger info "downloading kubernetes: $K8S_BIN_VER binaries"
   docker pull easzlab/kubeasz-k8s-bin:${K8S_BIN_VER} && \
-  echo "[INFO] run a temporary container" && \
+  logger debug "run a temporary container" && \
   docker run -d --name temp_k8s_bin easzlab/kubeasz-k8s-bin:${K8S_BIN_VER} && \
-  echo "[INFO] cp k8s binaries" && \
+  logger debug "cp k8s binaries" && \
-  docker cp temp_k8s_bin:/k8s /etc/ansible/k8s_bin_tmp && \
+  docker cp temp_k8s_bin:/k8s $BASE/k8s_bin_tmp && \
-  /bin/mv -f /etc/ansible/k8s_bin_tmp/* /etc/ansible/bin && \
+  /bin/mv -f $BASE/k8s_bin_tmp/* $BASE/bin && \
-  echo "[INFO] stop&remove temporary container" && \
+  logger debug "stop&remove temporary container" && \
   docker rm -f temp_k8s_bin && \
-  rm -rf /etc/ansible/k8s_bin_tmp
+  rm -rf $BASE/k8s_bin_tmp
 }
 
 function get_ext_bin() {
-  [[ -f "/etc/ansible/bin/etcdctl" ]] && { echo "[WARN] extral binaries existed"; return 0; }
+  [[ -f "$BASE/bin/etcdctl" ]] && { logger warn "extral binaries existed"; return 0; }
 
-  echo -e "[INFO] \033[33mdownloading extral binaries\033[0m kubeasz-ext-bin:$EXT_BIN_VER"
+  logger info "downloading extral binaries kubeasz-ext-bin:$EXT_BIN_VER"
   docker pull easzlab/kubeasz-ext-bin:${EXT_BIN_VER} && \
-  echo "[INFO] run a temporary container" && \
+  logger debug "run a temporary container" && \
   docker run -d --name temp_ext_bin easzlab/kubeasz-ext-bin:${EXT_BIN_VER} && \
-  echo "[INFO] cp extral binaries" && \
+  logger debug "cp extral binaries" && \
-  docker cp temp_ext_bin:/extra /etc/ansible/extra_bin_tmp && \
+  docker cp temp_ext_bin:/extra $BASE/extra_bin_tmp && \
-  /bin/mv -f /etc/ansible/extra_bin_tmp/* /etc/ansible/bin && \
+  /bin/mv -f $BASE/extra_bin_tmp/* $BASE/bin && \
-  echo "[INFO] stop&remove temporary container" && \
+  logger debug "stop&remove temporary container" && \
   docker rm -f temp_ext_bin && \
-  rm -rf /etc/ansible/extra_bin_tmp
+  rm -rf $BASE/extra_bin_tmp
 }
 
 function get_sys_pkg() {
-  [[ -f "/etc/ansible/down/packages/chrony_xenial.tar.gz" ]] && { echo "[WARN] system packages existed"; return 0; }
+  [[ -f "$BASE/down/packages/chrony_xenial.tar.gz" ]] && { logger warn "system packages existed"; return 0; }
 
-  echo -e "[INFO] \033[33mdownloading system packages\033[0m kubeasz-sys-pkg:$SYS_PKG_VER"
+  logger info "downloading system packages kubeasz-sys-pkg:$SYS_PKG_VER"
   docker pull easzlab/kubeasz-sys-pkg:${SYS_PKG_VER} && \
-  echo "[INFO] run a temporary container" && \
+  logger debug "run a temporary container" && \
   docker run -d --name temp_sys_pkg easzlab/kubeasz-sys-pkg:${SYS_PKG_VER} && \
-  echo "[INFO] cp system packages" && \
+  logger debug "cp system packages" && \
-  docker cp temp_sys_pkg:/packages /etc/ansible/down && \
+  docker cp temp_sys_pkg:/packages $BASE/down && \
-  echo "[INFO] stop&remove temporary container" && \
+  logger debug "stop&remove temporary container" && \
   docker rm -f temp_sys_pkg
 }
 
 function get_offline_image() {
 
-  imageDir=/etc/ansible/down
+  imageDir=$BASE/down
-  [[ -d "$imageDir" ]] || { echo "[ERROR] $imageDir not existed!"; exit 1; }
+  [[ -d "$imageDir" ]] || { logger error "$imageDir not existed!"; exit 1; }
   
-  echo -e "[INFO] \033[33mdownloading offline images\033[0m"
+  logger info "downloading offline images"
 
   if [[ ! -f "$imageDir/calico_$calicoVer.tar" ]];then
     docker pull "calico/cni:${calicoVer}" && \
@@ -239,6 +277,7 @@ function get_offline_image() {
 }
 
 function download_all() {
+  mkdir -p /opt/kube/bin "$BASE/down" "$BASE/bin"
   download_docker && \
   install_docker && \
   get_kubeasz && \
@@ -248,16 +287,17 @@ function download_all() {
 }
 
 function start_kubeasz_docker() {
-  [[ -d "/etc/ansible/roles/kube-node" ]] || { echo "[ERROR] not initialized. try 'easzup -D' first."; exit 1; }
+  [[ -d "$BASE/roles/kube-node" ]] || { logger error "not initialized. try 'ezdown -D' first."; exit 1; }
 
+  logger info "try to run kubeasz in a container"
   # get host's IP
   host_if=$(ip route|grep default|cut -d' ' -f5)
   host_ip=$(ip a|grep "$host_if$"|awk '{print $2}'|cut -d'/' -f1)
-  echo "[INFO] get host IP: $host_ip"
+  logger debug "get host IP: $host_ip"
 
   # allow ssh login using key locally
   if [[ ! -e /root/.ssh/id_rsa ]]; then
-    echo "[INFO] generate ssh key pair"
+    logger debug "generate ssh key pair"
     ssh-keygen -t rsa -b 2048 -N '' -f /root/.ssh/id_rsa > /dev/null
     cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys
     ssh-keyscan -t ecdsa -H "$host_ip" >> /root/.ssh/known_hosts
@@ -265,20 +305,19 @@ function start_kubeasz_docker() {
 
   # create a link '/usr/bin/python' in Ubuntu1604
   if [[ ! -e /usr/bin/python && -e /etc/debian_version ]]; then
-    echo "[INFO] create a soft link '/usr/bin/python'"
+    logger debug "create a soft link '/usr/bin/python'"
     ln -s /usr/bin/python3 /usr/bin/python
   fi
 
   # 
-  docker load -i /etc/ansible/down/kubeasz_${KUBEASZ_VER}.tar
+  docker load -i $BASE/down/kubeasz_${KUBEASZ_VER}.tar
 
   # run kubeasz docker container
-  echo "[INFO] run kubeasz in a container"
   docker run --detach \
       --name kubeasz \
       --restart always \
       --env HOST_IP="$host_ip" \
-      --volume /etc/ansible:/etc/ansible \
+      --volume $BASE:$BASE \
       --volume /root/.kube:/root/.kube \
       --volume /root/.ssh/id_rsa:/root/.ssh/id_rsa:ro \
       --volume /root/.ssh/id_rsa.pub:/root/.ssh/id_rsa.pub:ro \
@@ -287,35 +326,19 @@ function start_kubeasz_docker() {
 }
 
 function clean_container() {
- echo "[INFO] clean all running containers"
+ logger info "clean all running containers"
  docker ps -a|awk 'NR>1{print $1}'|xargs docker rm -f 
 } 
 
-function usage() {
-  cat <<EOF
-Usage: easzup [options] [args]
-  option: -{DdekSz}
-    -C         stop&clean all local containers
-    -D         download all into /etc/ansible
-    -P         download system packages for offline installing
-    -S         start kubeasz in a container
-    -d <ver>   set docker-ce version, default "$DOCKER_VER"
-    -e <ver>   set kubeasz-ext-bin version, default "$EXT_BIN_VER"
-    -k <ver>   set kubeasz-k8s-bin version, default "$K8S_BIN_VER"
-    -m <str>   set docker registry mirrors, default "CN"(used in Mainland,China)
-    -p <ver>   set kubeasz-sys-pkg version, default "$SYS_PKG_VER"
-    -z <ver>   set kubeasz version, default "$KUBEASZ_VER"
-  
-see more at https://github.com/kubeasz/dockerfiles
-EOF
-} 
 
 ### Main Lines ##################################################
 function main() {
+  BASE="/etc/kubeasz"
+
   # check if use bash shell
-  readlink /proc/$$/exe|grep -q "dash" && { echo "[ERROR] you should use bash shell, not sh"; exit 1; }
+  readlink /proc/$$/exe|grep -q "dash" && { logger error "you should use bash shell, not sh"; exit 1; }
   # check if use with root
-  [[ "$EUID" -ne 0 ]] && { echo "[ERROR] you should run this script as root"; exit 1; }
+  [[ "$EUID" -ne 0 ]] && { logger error "you should run this script as root"; exit 1; }
   
   [[ "$#" -eq 0 ]] && { usage >&2; exit 1; }
   
@@ -360,12 +383,12 @@ function main() {
       esac
   done
   
-  [[ "$ACTION" == "" ]] && { echo "[ERROR] illegal option"; usage; exit 1; }
+  [[ "$ACTION" == "" ]] && { logger error "illegal option"; usage; exit 1; }
   
   # excute cmd "$ACTION" 
-  echo -e "[INFO] \033[33mAction begin\033[0m : $ACTION"
+  logger info "Action begin: $ACTION"
-  ${ACTION} || { echo -e "[ERROR] \033[31mAction failed\033[0m : $ACTION"; return 1; }
+  ${ACTION} || { logger error "Action failed: $ACTION"; return 1; }
-  echo -e "[INFO] \033[32mAction successed\033[0m : $ACTION"
+  logger info "Action successed: $ACTION"
 }
 
 main "$@"

playbooks/21.addetcd.yml

@@ -12,9 +12,9 @@
       shell: 'for ip in {{ NODE_IPS }};do \
                 ETCDCTL_API=3 {{ base_dir }}/bin/etcdctl \
                 --endpoints=https://"$ip":2379 \
-                --cacert={{ base_dir }}/.cluster/ssl/ca.pem \
+                --cacert={{ cluster_dir }}/ssl/ca.pem \
-                --cert={{ base_dir }}/.cluster/ssl/admin.pem \
+                --cert={{ cluster_dir }}/ssl/admin.pem \
-                --key={{ base_dir }}/.cluster/ssl/admin-key.pem \
+                --key={{ cluster_dir }}/ssl/admin-key.pem \
                 endpoint health; \
               done'
       register: ETCD_CLUSTER_STATUS

playbooks/31.deletcd.yml

@@ -24,9 +24,9 @@
       shell: 'for ip in {{ NODE_IPS }};do \
                 ETCDCTL_API=3 {{ base_dir }}/bin/etcdctl \
                 --endpoints=https://"$ip":2379 \
-                --cacert={{ base_dir }}/.cluster/ssl/ca.pem \
+                --cacert={{ cluster_dir }}/ssl/ca.pem \
-                --cert={{ base_dir }}/.cluster/ssl/admin.pem \
+                --cert={{ cluster_dir }}/ssl/admin.pem \
-                --key={{ base_dir }}/.cluster/ssl/admin-key.pem \
+                --key={{ cluster_dir }}/ssl/admin-key.pem \
                 endpoint health; \
               done'
       register: ETCD_CLUSTER_STATUS

playbooks/93.upgrade.yml

@@ -1,7 +1,6 @@
 # WARNING: Upgrade the k8s cluster can be risky. Make sure you know what you are doing.
 # Read the guide: 'op/upgrade.md' .
-# Usage: `ansible-playbook /etc/ansible/22.upgrade.yml -t upgrade_k8s`
+# Usage: ezctl <cluster_name> upgrade 
-#	or `easzctl upgrade`
 
 # update masters
 - hosts: 

playbooks/94.backup.yml

@@ -12,9 +12,9 @@
     shell: 'for ip in {{ NODE_IPS }};do \
               ETCDCTL_API=3 {{ base_dir }}/bin/etcdctl \
               --endpoints=https://"$ip":2379 \
-              --cacert={{ base_dir }}/.cluster/ssl/ca.pem \
+              --cacert={{ cluster_dir }}/ssl/ca.pem \
-              --cert={{ base_dir }}/.cluster/ssl/admin.pem \
+              --cert={{ cluster_dir }}/ssl/admin.pem \
-              --key={{ base_dir }}/.cluster/ssl/admin-key.pem \
+              --key={{ cluster_dir }}/ssl/admin-key.pem \
               endpoint health; \
             done'
     register: ETCD_CLUSTER_STATUS
@@ -42,7 +42,7 @@
   - name: fetch the backup data
     fetch:
       src: /etcd_backup/snapshot_{{ temp }}.db
-      dest: "{{ base_dir }}/.cluster/backup/"
+      dest: "{{ cluster_dir }}/backup/"
       flat: yes
     delegate_to: "{{ RUNNING_NODE.stdout }}"
 
@@ -52,14 +52,14 @@
   - name: Backing up ansible hosts-1
     copy:
       src: "{{ base_dir }}/hosts"
-      dest: "{{ base_dir }}/.cluster/backup/hosts"
+      dest: "{{ cluster_dir }}/backup/hosts"
     register: p
 
   - name: Backing up ansible hosts-2
-    shell: "cd {{ base_dir }}/.cluster/backup && \
+    shell: "cd {{ cluster_dir }}/backup && \
 	cp -fp hosts hosts-$(date +'%Y%m%d%H%M')"
     when: 'p is changed'
 
   #- name: Backing up etcd snapshot with datetime
-  #  shell: "cd {{ base_dir }}/.cluster/backup && \
+  #  shell: "cd {{ cluster_dir }}/backup && \
   #	cp -fp snapshot.db snapshot-$(date +'%Y%m%d%H%M').db"

roles/calico/defaults/main.yml

@@ -1,28 +0,0 @@
-# 部分calico相关配置，更全配置可以去roles/calico/templates/calico.yaml.j2自定义
-
-# etcd 集群服务地址列表, 根据etcd组成员自动生成
-TMP_ENDPOINTS: "{% for h in groups['etcd'] %}https://{{ h }}:2379,{% endfor %}"
-ETCD_ENDPOINTS: "{{ TMP_ENDPOINTS.rstrip(',') }}"
-
-# 设置 CALICO_IPV4POOL_IPIP=“off”,可以提高网络性能，条件限制详见 docs/setup/calico.md
-CALICO_IPV4POOL_IPIP: "Always"
-
-# 设置 Felix 日志级别
-FELIX_LOG_LVL: "warning"
-
-# 设置 calico-node使用的host IP，bgp邻居通过该地址建立，可手工指定也可以自动发现
-#IP_AUTODETECTION_METHOD: "interface=eth0"
-IP_AUTODETECTION_METHOD: "can-reach={{ groups['kube-master'][0] }}"
-
-# 设置calico 网络 backend: brid, vxlan, none
-CALICO_NETWORKING_BACKEND: "brid"
-
-# 更新支持calico 版本: [v3.3.x] [v3.4.x] [v3.8.x] [v3.15.x]
-calicoVer: "v3.15.3"
-calico_ver: "{{ calicoVer }}"
-
-# calico 主版本
-calico_ver_main: "{{ calico_ver.split('.')[0] }}.{{ calico_ver.split('.')[1] }}"
-
-# 离线镜像tar包
-calico_offline: "calico_{{ calico_ver }}.tar"

roles/calico/templates/calico-v3.15.yaml.j2

@@ -344,7 +344,7 @@ spec:
               value: "false"
             # Set Felix logging to "info"
             - name: FELIX_LOGSEVERITYSCREEN
-              value: "info"
+              value: "warning"
             - name: FELIX_HEALTHENABLED
               value: "true"
             # Set Kubernetes NodePorts: If services do use NodePorts outside Calico’s expected range,

roles/calico/templates/calico-v3.3.yaml.j2

@@ -180,7 +180,7 @@ spec:
               value: "false"
             # Set Felix logging 
             - name: FELIX_LOGSEVERITYSCREEN
-              value: "{{ FELIX_LOG_LVL }}"
+              value: "warning"
             - name: FELIX_HEALTHENABLED
               value: "true"
             # Set Kubernetes NodePorts: If services do use NodePorts outside Calico’s expected range,

roles/calico/templates/calico-v3.4.yaml.j2

@@ -218,7 +218,7 @@ spec:
               value: "false"
             # Set Felix logging 
             - name: FELIX_LOGSEVERITYSCREEN
-              value: "{{ FELIX_LOG_LVL }}"
+              value: "warning"
             - name: FELIX_HEALTHENABLED
               value: "true"
             # Set Kubernetes NodePorts: If services do use NodePorts outside Calico’s expected range,

roles/calico/templates/calico-v3.8.yaml.j2

@@ -318,7 +318,7 @@ spec:
               value: "false"
             # Set Felix logging to "info"
             - name: FELIX_LOGSEVERITYSCREEN
-              value: "{{ FELIX_LOG_LVL }}"
+              value: "warning"
             - name: FELIX_HEALTHENABLED
               value: "true"
             # Set Kubernetes NodePorts: If services do use NodePorts outside Calico’s expected range,

roles/calico/vars/main.yml

@@ -0,0 +1,3 @@
+# etcd 集群服务地址列表, 根据etcd组成员自动生成
+TMP_ENDPOINTS: "{% for h in groups['etcd'] %}https://{{ h }}:2379,{% endfor %}"
+ETCD_ENDPOINTS: "{{ TMP_ENDPOINTS.rstrip(',') }}"

roles/chrony/defaults/main.yml

@@ -1,13 +0,0 @@
-# 设置时间源服务器
-ntp_servers:
-  - "ntp1.aliyun.com"
-  - "ntp2.aliyun.com"
-  - "time1.cloud.tencent.com"
-  - "time2.cloud.tencent.com"
-  - "0.cn.pool.ntp.org"
-
-# 设置允许内部时间同步的网络段，比如"10.0.0.0/8"，默认全部允许
-local_network: "0.0.0.0/0"
-
-# 离线安装 chrony (offline|online)
-INSTALL_SOURCE: "online"

roles/cilium/defaults/main.yml

@@ -1,14 +0,0 @@
-# 部分cilium相关配置, Note: cilium 需要Linux kernel >= 4.9.17
-
-# 如果 node 节点有多块网卡，请设置 true
-# 如果发现‘dial tcp 10.68.0.1:443: i/o timeout’的错误，请设置 true
-NODE_WITH_MULTIPLE_NETWORKS: "true"
-
-# 镜像版本
-cilium_ver: "v1.4.1"
-
-# 离线镜像tar包
-cilium_offline: "cilium_{{ cilium_ver }}.tar"
-
-# CILIUM_ETCD_OPERATOR 创建的 etcd 集群节点数 1,3,5,7...
-ETCD_CLUSTER_SIZE: 1

roles/cluster-addon/vars/main.yml

@@ -1,10 +1,8 @@
-# dns 自动安装，'dns_backend'可选"coredns"和“kubedns”
-dns_install: "yes"
-dns_backend: "coredns"
 # 设置 dns svc ip (这里选用 SERVICE_CIDR 中第2个IP)
 CLUSTER_DNS_SVC_IP: "{{ SERVICE_CIDR | ipaddr('net') | ipaddr(2) | ipaddr('address') }}"
+
+#
 kubednsVer: "1.14.13"
-corednsVer: "1.7.1"
 kubedns_offline: "kubedns_{{ kubednsVer }}.tar"
 coredns_offline: "coredns_{{ corednsVer }}.tar"
 dns_offline: "{%- if dns_backend == 'coredns' -%} \
@@ -13,17 +11,9 @@ dns_offline: "{%- if dns_backend == 'coredns' -%} \
                 {{ kubedns_offline }} \
               {%- endif -%}"
 
-# metric server 自动安装
-metricsserver_install: "yes"
-metricsVer: "v0.3.6"
 metricsserver_offline: "metrics-server_{{ metricsVer }}.tar"
 
-# dashboard 自动安装
-# dashboard v2.x.x 不依赖于heapster
-dashboard_install: "yes"
-dashboardVer: "v2.0.4"
 dashboard_offline: "dashboard_{{ dashboardVer }}.tar"
-dashboardMetricsScraperVer: "v1.0.4"
 metricsscraper_offline: "metrics-scraper_{{ dashboardMetricsScraperVer }}.tar"
 
 # ingress 自动安装，可选 "traefik" 和 "nginx-ingress"
@@ -36,9 +26,3 @@ metricsscraper_offline: "metrics-scraper_{{ dashboardMetricsScraperVer }}.tar"
 #metallb_protocol: "layer2"
 #metallb_offline: "metallb_{{ metallbVer }}.tar"
 #metallb_vip_pool: "192.168.1.240/29"
-
-# efk 自动安装
-#efk_install: "no"
-
-# prometheus 自动安装
-#prometheus_install: "no"

roles/cluster-restore/defaults/main.yml

@@ -1,5 +1,5 @@
 # 指定需要恢复的 etcd 数据备份，默认使用最近的一次备份
-# 在ansible 控制端查看备份目录：/etc/ansible/.cluster/backup
+# 在ansible 控制端查看备份目录：/etc/kubeasz/clusters/_cluster_name_/backup
 db_to_restore: "snapshot.db"
 
 # etcd 集群间通信的IP和端口, 根据etcd组成员自动生成

roles/cluster-restore/tasks/main.yml

@@ -9,7 +9,7 @@
 
 - name: 准备指定的备份etcd 数据
   copy: 
-    src: "{{ base_dir }}/.cluster/backup/{{ db_to_restore }}"
+    src: "{{ cluster_dir }}/backup/{{ db_to_restore }}"
     dest: "/etcd_backup/snapshot.db"
 
 - name: 清理上次备份恢复数据

roles/containerd/defaults/main.yml

@@ -1,8 +0,0 @@
-# 容器持久化存储目录
-STORAGE_DIR: "/var/lib/containerd"
-
-# 基础容器镜像
-SANDBOX_IMAGE: "easzlab/pause-amd64:3.2"
-
-# 启用容器仓库镜像
-ENABLE_MIRROR_REGISTRY: true

roles/containerd/templates/config.toml.j2

@@ -1,4 +1,4 @@
-root = "{{ STORAGE_DIR }}"
+root = "{{ CONTAINERD_STORAGE_DIR }}"
 state = "/run/containerd"
 oom_score = -999
 

roles/deploy/defaults/main.yml

@@ -1,13 +0,0 @@
-# CA 证书相关参数
-CA_EXPIRY: "876000h"
-CERT_EXPIRY: "438000h"
-
-# apiserver 默认第一个master节点
-KUBE_APISERVER: "https://{{ groups['kube-master'][0] }}:6443"
-
-# kubeconfig 配置参数，注意权限根据‘USER_NAME’设置：
-# 'admin' 表示创建集群管理员（所有）权限的 kubeconfig 
-# 'read' 表示创建只读权限的 kubeconfig
-CLUSTER_NAME: "cluster1"
-USER_NAME: "admin"
-CONTEXT_NAME: "context-{{ CLUSTER_NAME }}-{{ USER_NAME }}"

roles/deploy/tasks/create-kube-controller-manager-kubeconfig.yml

@@ -1,8 +1,8 @@
 - name: 准备kube-controller-manager 证书签名请求
-  template: src=kube-controller-manager-csr.json.j2 dest={{ base_dir }}/.cluster/ssl/kube-controller-manager-csr.json
+  template: src=kube-controller-manager-csr.json.j2 dest={{ cluster_dir }}/ssl/kube-controller-manager-csr.json
 
 - name: 创建 kube-controller-manager证书与私钥
-  shell: "cd {{ base_dir }}/.cluster/ssl && {{ base_dir }}/bin/cfssl gencert \
+  shell: "cd {{ cluster_dir }}/ssl && {{ base_dir }}/bin/cfssl gencert \
         -ca=ca.pem \
         -ca-key=ca-key.pem \
         -config=ca-config.json \
@@ -10,24 +10,24 @@
 
 - name: 设置集群参数
   shell: "{{ base_dir }}/bin/kubectl config set-cluster kubernetes \
-        --certificate-authority={{ base_dir }}/.cluster/ssl/ca.pem \
+        --certificate-authority={{ cluster_dir }}/ssl/ca.pem \
         --embed-certs=true \
         --server={{ KUBE_APISERVER }} \
-        --kubeconfig={{ base_dir }}/.cluster/kube-controller-manager.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-controller-manager.kubeconfig"
 
 - name: 设置认证参数
   shell: "{{ base_dir }}/bin/kubectl config set-credentials system:kube-controller-manager \
-        --client-certificate={{ base_dir }}/.cluster/ssl/kube-controller-manager.pem \
+        --client-certificate={{ cluster_dir }}/ssl/kube-controller-manager.pem \
-        --client-key={{ base_dir }}/.cluster/ssl/kube-controller-manager-key.pem \
+        --client-key={{ cluster_dir }}/ssl/kube-controller-manager-key.pem \
         --embed-certs=true \
-        --kubeconfig={{ base_dir }}/.cluster/kube-controller-manager.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-controller-manager.kubeconfig"
 
 - name: 设置上下文参数
   shell: "{{ base_dir }}/bin/kubectl config set-context default \
         --cluster=kubernetes \
         --user=system:kube-controller-manager \
-        --kubeconfig={{ base_dir }}/.cluster/kube-controller-manager.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-controller-manager.kubeconfig"
 
 - name: 选择默认上下文
   shell: "{{ base_dir }}/bin/kubectl config use-context default \
-   --kubeconfig={{ base_dir }}/.cluster/kube-controller-manager.kubeconfig"
+   --kubeconfig={{ cluster_dir }}/kube-controller-manager.kubeconfig"

roles/deploy/tasks/create-kube-proxy-kubeconfig.yml

@@ -1,8 +1,8 @@
 - name: 准备kube-proxy 证书签名请求
-  template: src=kube-proxy-csr.json.j2 dest={{ base_dir }}/.cluster/ssl/kube-proxy-csr.json
+  template: src=kube-proxy-csr.json.j2 dest={{ cluster_dir }}/ssl/kube-proxy-csr.json
 
 - name: 创建 kube-proxy证书与私钥
-  shell: "cd {{ base_dir }}/.cluster/ssl && {{ base_dir }}/bin/cfssl gencert \
+  shell: "cd {{ cluster_dir }}/ssl && {{ base_dir }}/bin/cfssl gencert \
         -ca=ca.pem \
         -ca-key=ca-key.pem \
         -config=ca-config.json \
@@ -10,24 +10,24 @@
 
 - name: 设置集群参数
   shell: "{{ base_dir }}/bin/kubectl config set-cluster kubernetes \
-        --certificate-authority={{ base_dir }}/.cluster/ssl/ca.pem \
+        --certificate-authority={{ cluster_dir }}/ssl/ca.pem \
         --embed-certs=true \
         --server={{ KUBE_APISERVER }} \
-        --kubeconfig={{ base_dir }}/.cluster/kube-proxy.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-proxy.kubeconfig"
 
 - name: 设置客户端认证参数
   shell: "{{ base_dir }}/bin/kubectl config set-credentials kube-proxy \
-        --client-certificate={{ base_dir }}/.cluster/ssl/kube-proxy.pem \
+        --client-certificate={{ cluster_dir }}/ssl/kube-proxy.pem \
-        --client-key={{ base_dir }}/.cluster/ssl/kube-proxy-key.pem \
+        --client-key={{ cluster_dir }}/ssl/kube-proxy-key.pem \
         --embed-certs=true \
-        --kubeconfig={{ base_dir }}/.cluster/kube-proxy.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-proxy.kubeconfig"
 
 - name: 设置上下文参数
   shell: "{{ base_dir }}/bin/kubectl config set-context default \
         --cluster=kubernetes \
         --user=kube-proxy \
-        --kubeconfig={{ base_dir }}/.cluster/kube-proxy.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-proxy.kubeconfig"
 
 - name: 选择默认上下文
   shell: "{{ base_dir }}/bin/kubectl config use-context default \
-   --kubeconfig={{ base_dir }}/.cluster/kube-proxy.kubeconfig"
+   --kubeconfig={{ cluster_dir }}/kube-proxy.kubeconfig"

roles/deploy/tasks/create-kube-scheduler-kubeconfig.yml

@@ -1,8 +1,8 @@
 - name: 准备kube-scheduler 证书签名请求
-  template: src=kube-scheduler-csr.json.j2 dest={{ base_dir }}/.cluster/ssl/kube-scheduler-csr.json
+  template: src=kube-scheduler-csr.json.j2 dest={{ cluster_dir }}/ssl/kube-scheduler-csr.json
 
 - name: 创建 kube-scheduler证书与私钥
-  shell: "cd {{ base_dir }}/.cluster/ssl && {{ base_dir }}/bin/cfssl gencert \
+  shell: "cd {{ cluster_dir }}/ssl && {{ base_dir }}/bin/cfssl gencert \
         -ca=ca.pem \
         -ca-key=ca-key.pem \
         -config=ca-config.json \
@@ -10,24 +10,24 @@
 
 - name: 设置集群参数
   shell: "{{ base_dir }}/bin/kubectl config set-cluster kubernetes \
-        --certificate-authority={{ base_dir }}/.cluster/ssl/ca.pem \
+        --certificate-authority={{ cluster_dir }}/ssl/ca.pem \
         --embed-certs=true \
         --server={{ KUBE_APISERVER }} \
-        --kubeconfig={{ base_dir }}/.cluster/kube-scheduler.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-scheduler.kubeconfig"
 
 - name: 设置认证参数
   shell: "{{ base_dir }}/bin/kubectl config set-credentials system:kube-scheduler \
-        --client-certificate={{ base_dir }}/.cluster/ssl/kube-scheduler.pem \
+        --client-certificate={{ cluster_dir }}/ssl/kube-scheduler.pem \
-        --client-key={{ base_dir }}/.cluster/ssl/kube-scheduler-key.pem \
+        --client-key={{ cluster_dir }}/ssl/kube-scheduler-key.pem \
         --embed-certs=true \
-        --kubeconfig={{ base_dir }}/.cluster/kube-scheduler.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-scheduler.kubeconfig"
 
 - name: 设置上下文参数
   shell: "{{ base_dir }}/bin/kubectl config set-context default \
         --cluster=kubernetes \
         --user=system:kube-scheduler \
-        --kubeconfig={{ base_dir }}/.cluster/kube-scheduler.kubeconfig"
+        --kubeconfig={{ cluster_dir }}/kube-scheduler.kubeconfig"
 
 - name: 选择默认上下文
   shell: "{{ base_dir }}/bin/kubectl config use-context default \
-   --kubeconfig={{ base_dir }}/.cluster/kube-scheduler.kubeconfig"
+   --kubeconfig={{ cluster_dir }}/kube-scheduler.kubeconfig"

roles/deploy/tasks/create-kubectl-kubeconfig.yml

@@ -11,10 +11,10 @@
   when: USER_NAME == "read"
 
 - name: 准备kubectl使用的{{ USER_NAME }}证书签名请求
-  template: src={{ USER_NAME }}-csr.json.j2 dest={{ base_dir }}/.cluster/ssl/{{ USER_NAME }}-csr.json
+  template: src={{ USER_NAME }}-csr.json.j2 dest={{ cluster_dir }}/ssl/{{ USER_NAME }}-csr.json
 
 - name: 创建{{ USER_NAME }}证书与私钥
-  shell: "cd {{ base_dir }}/.cluster/ssl && {{ base_dir }}/bin/cfssl gencert \
+  shell: "cd {{ cluster_dir }}/ssl && {{ base_dir }}/bin/cfssl gencert \
         -ca=ca.pem \
         -ca-key=ca-key.pem \
         -config=ca-config.json \
@@ -22,15 +22,15 @@
 
 - name: 设置集群参数
   shell: "{{ base_dir }}/bin/kubectl config set-cluster {{ CLUSTER_NAME }} \
-        --certificate-authority={{ base_dir }}/.cluster/ssl/ca.pem \
+        --certificate-authority={{ cluster_dir }}/ssl/ca.pem \
         --embed-certs=true \
         --server={{ KUBE_APISERVER }}"
 
 - name: 设置客户端认证参数
   shell: "{{ base_dir }}/bin/kubectl config set-credentials {{ USER_NAME }} \
-        --client-certificate={{ base_dir }}/.cluster/ssl/{{ USER_NAME }}.pem \
+        --client-certificate={{ cluster_dir }}/ssl/{{ USER_NAME }}.pem \
         --embed-certs=true \
-        --client-key={{ base_dir }}/.cluster/ssl/{{ USER_NAME }}-key.pem"
+        --client-key={{ cluster_dir }}/ssl/{{ USER_NAME }}-key.pem"
 
 - name: 设置上下文参数
   shell: "{{ base_dir }}/bin/kubectl config set-context {{ CONTEXT_NAME }} \

roles/deploy/tasks/main.yml

@@ -1,8 +1,8 @@
 - name: prepare some dirs
   file: name={{ item }} state=directory
   with_items:
-  - "{{ base_dir }}/.cluster/ssl"
+  - "{{ cluster_dir }}/ssl"
-  - "{{ base_dir }}/.cluster/backup"
+  - "{{ cluster_dir }}/backup"
 
 - name: 本地设置 bin 目录权限
   file: path={{ base_dir }}/bin state=directory mode=0755 recurse=yes
@@ -10,11 +10,11 @@
 # 注册变量p，根据p的stat信息判断是否已经生成过ca证书，如果没有，下一步生成证书
 # 如果已经有ca证书，为了保证整个安装的幂等性，跳过证书生成的步骤
 - name: 读取ca证书stat信息
-  stat: path="{{ base_dir }}/.cluster/ssl/ca.pem"
+  stat: path="{{ cluster_dir }}/ssl/ca.pem"
   register: p
 
 - name: 准备CA配置文件和签名请求
-  template: src={{ item }}.j2 dest={{ base_dir }}/.cluster/ssl/{{ item }}
+  template: src={{ item }}.j2 dest={{ cluster_dir }}/ssl/{{ item }}
   with_items:
   - "ca-config.json"
   - "ca-csr.json"
@@ -22,7 +22,7 @@
 
 - name: 生成 CA 证书和私钥
   when: p.stat.isreg is not defined
-  shell: "cd {{ base_dir }}/.cluster/ssl && \
+  shell: "cd {{ cluster_dir }}/ssl && \
 	 {{ base_dir }}/bin/cfssl gencert -initca ca-csr.json | {{ base_dir }}/bin/cfssljson -bare ca" 
 
 #----------- 创建配置文件: /root/.kube/config

roles/deploy/vars/main.yml

@@ -0,0 +1,2 @@
+# apiserver 默认第一个master节点
+KUBE_APISERVER: "https://{{ groups['kube-master'][0] }}:6443"

roles/docker/defaults/main.yml

@@ -1,14 +0,0 @@
-# docker容器存储目录
-STORAGE_DIR: "/var/lib/docker"
-
-# 开启Restful API
-ENABLE_REMOTE_API: false
-
-# 启用 docker 仓库镜像
-ENABLE_MIRROR_REGISTRY: true
-
-# 设置 docker 仓库镜像
-REG_MIRRORS: '["https://docker.mirrors.ustc.edu.cn", "http://hub-mirror.c.163.com"]'
-
-# 信任的HTTP仓库
-INSECURE_REG: '["127.0.0.1/8"]'

roles/docker/templates/daemon.json.j2

@@ -1,8 +1,11 @@
 {
-  "data-root": "{{ STORAGE_DIR }}",
+  "data-root": "{{ DOCKER_STORAGE_DIR }}",
   "exec-opts": ["native.cgroupdriver=cgroupfs"],
 {% if ENABLE_MIRROR_REGISTRY %}
-  "registry-mirrors": {{ REG_MIRRORS }}, 
+  "registry-mirrors": [
+    "https://docker.mirrors.ustc.edu.cn",
+    "http://hub-mirror.c.163.com"
+  ], 
 {% endif %}
 {% if ENABLE_REMOTE_API %}
   "hosts": ["tcp://0.0.0.0:2376", "unix:///var/run/docker.sock"],

roles/etcd/tasks/main.yml

@@ -14,7 +14,7 @@
   tags: upgrade_etcd
 
 - name: 分发证书相关
-  copy: src={{ base_dir }}/.cluster/ssl/{{ item }} dest={{ ca_dir }}/{{ item }}
+  copy: src={{ cluster_dir }}/ssl/{{ item }} dest={{ ca_dir }}/{{ item }}
   with_items:
   - ca.pem
   - ca-key.pem

roles/flannel/defaults/main.yml

@@ -1,13 +0,0 @@
-# 部分flannel配置，参考 docs/setup/network-plugin/flannel.md
-
-# 设置flannel 后端
-#FLANNEL_BACKEND: "host-gw"
-FLANNEL_BACKEND: "vxlan"
-DIRECT_ROUTING: false
-
-#flanneld_image: "quay.io/coreos/flannel:v0.10.0-amd64"
-flannelVer: "v0.13.0-amd64"
-flanneld_image: "easzlab/flannel:{{ flannelVer }}"
-
-# 离线镜像tar包
-flannel_offline: "flannel_{{ flannelVer }}.tar"

roles/harbor/tasks/main.yml

@@ -34,7 +34,7 @@
   
   - block:
     - name: 生成自签名证书相关
-      copy: src={{ base_dir }}/.cluster/ssl/{{ item }} dest={{ ca_dir }}/{{ item }}
+      copy: src={{ cluster_dir }}/ssl/{{ item }} dest={{ ca_dir }}/{{ item }}
       with_items:
       - ca.pem
       - ca-key.pem

roles/harbor/vars/main.yml

@@ -1,9 +1,6 @@
-# harbor version，完整版本号，目前支持 v1.5.x , v1.6.x, v1.7.x, v1.8.x, v1.9.x, v1.10.x
-HARBOR_VER: "v1.9.4"
-
 # harbor 主版本号，目前支持主版本号 v1.5/v1.6/v1.7/v1.8/v1.9/v.10
 # 从完整版本号提取出主版本号 v1.5/v1.6/v1.7/v1.8/v1.9/v.10
 HARBOR_VER_MAIN: "{{ HARBOR_VER.split('.')[0] }}.{{ HARBOR_VER.split('.')[1] }}"
 
 # HARBOR_HOSTNAME 值设置
-HARBOR_HOSTNAME: "{{ inventory_hostname if (HARBOR_DOMAIN == '') else HARBOR_DOMAIN }}"
+HARBOR_HOSTNAME: "{% if HARBOR_DOMAIN != '' %}HARBOR_DOMAIN{% else %}inventory_hostname{% endif %}" 

roles/kube-master/defaults/main.yml

@@ -1,29 +0,0 @@
-# etcd 集群服务地址列表, 根据etcd组成员自动生成
-TMP_ENDPOINTS: "{% for h in groups['etcd'] %}https://{{ h }}:2379,{% endfor %}"
-ETCD_ENDPOINTS: "{{ TMP_ENDPOINTS.rstrip(',') }}" 
-# 设置 dns svc ip (这里选用 SERVICE_CIDR 中第1个IP)
-CLUSTER_KUBERNETES_SVC_IP: "{{ SERVICE_CIDR | ipaddr('net') | ipaddr(1) | ipaddr('address') }}"
-# k8s 集群 master 节点证书配置，可以添加多个ip和域名（比如增加公网ip和域名）
-MASTER_CERT_HOSTS:
-  - "10.1.1.1"
-  - "k8s.test.io"
-  #- "61.182.11.41"
-  #- "www.test.com"
-
-# apiserver 基础认证（用户名/密码）配置，详见 master 节点文件‘/etc/kubernetes/ssl/basic-auth.csv’
-# the first three values can be anything;
-# These tokens are arbitrary but should represent at least 128 bits of entropy derived from
-# a secure random number generator, for example:
-# head -c 16 /dev/urandom | od -An -t x | tr -d ' '
-# 例子： 02b50b05283e98dd0fd71db496ef01e8,kubelet-bootstrap,10001,"system:bootstrappers"
-# 是否启用基础认证 yes/no
-BASIC_AUTH_ENABLE: 'yes'
-# 用户名：
-BASIC_AUTH_USER: 'admin'
-# 密码：初次运行时会生成随机密码
-BASIC_AUTH_PASS: '92c068405aa491239b56140ea6b3b44b'
-
-# node 节点上 pod 网段掩码长度（决定每个节点最多能分配的pod ip地址）
-# 如果flannel 使用 --kube-subnet-mgr 参数，那么它将读取该设置为每个节点分配pod网段
-# https://github.com/coreos/flannel/issues/847
-NODE_CIDR_LEN: 24

roles/kube-master/tasks/main.yml

@@ -30,25 +30,6 @@
         -config={{ ca_dir }}/ca-config.json \
         -profile=kubernetes aggregator-proxy-csr.json | {{ bin_dir }}/cfssljson -bare aggregator-proxy"
 
-- block:
-  - name: 生成 basic-auth 随机密码
-    shell: 'PWD=`head -c 16 /dev/urandom | od -An -t x | tr -d " "`; \
-	sed -i "s/_pwd_/$PWD/g" {{ base_dir }}/roles/kube-master/defaults/main.yml; \
-	echo $PWD;'
-    connection: local
-    register: TMP_PASS
-    run_once: true
-    
-  - name: 设置 basic-auth 随机密码
-    set_fact: BASIC_AUTH_PASS={{ TMP_PASS.stdout }}
-  when: 'BASIC_AUTH_ENABLE == "yes" and BASIC_AUTH_PASS == "_pwd_"'
-  tags: restart_master
-  
-- name: 创建 basic-auth.csv
-  template: src=basic-auth.csv.j2 dest={{ ca_dir }}/basic-auth.csv
-  when: 'BASIC_AUTH_ENABLE == "yes"'
-  tags: restart_master
-
 - name: 替换 kubeconfig 的 apiserver 地址
   lineinfile:
     dest: "{{ item }}"
@@ -84,14 +65,3 @@
   retries:  5
   delay: 6
   tags: upgrade_k8s, restart_master
-
-- name: 配置{{ BASIC_AUTH_USER }}用户rbac权限
-  template: src=basic-auth-rbac.yaml.j2 dest=/opt/kube/basic-auth-rbac.yaml
-  when: 'BASIC_AUTH_ENABLE == "yes"'
-  tags: restart_master
-
-- name: 创建{{ BASIC_AUTH_USER }}用户rbac权限
-  shell: "{{ bin_dir }}/kubectl apply -f /opt/kube/basic-auth-rbac.yaml"
-  when: 'BASIC_AUTH_ENABLE == "yes"'
-  run_once: true
-  tags: restart_master

roles/kube-master/templates/basic-auth-rbac.yaml.j2

@@ -1,12 +0,0 @@
-apiVersion: rbac.authorization.k8s.io/v1
-kind: ClusterRoleBinding
-metadata:
-  name: basic-auth-binding
-roleRef:
-  apiGroup: rbac.authorization.k8s.io
-  kind: ClusterRole
-  name: cluster-admin
-subjects:
-- apiGroup: rbac.authorization.k8s.io
-  kind: User
-  name: {{ BASIC_AUTH_USER }}

roles/kube-master/templates/basic-auth.csv.j2

@@ -1,2 +0,0 @@
-{{ BASIC_AUTH_PASS }},{{ BASIC_AUTH_USER }},1
-{{ BASIC_AUTH_PASS | truncate(8, True, '') }},readonly,2

roles/kube-master/templates/kube-apiserver.service.j2

@@ -10,9 +10,6 @@ ExecStart={{ bin_dir }}/kube-apiserver \
   --anonymous-auth=false \
   --api-audiences=api,istio-ca \
   --authorization-mode=Node,RBAC \
-{% if BASIC_AUTH_ENABLE == "yes" %}
-  --token-auth-file={{ ca_dir }}/basic-auth.csv \
-{% endif %}
   --bind-address={{ inventory_hostname }} \
   --client-ca-file={{ ca_dir }}/ca.pem \
   --endpoint-reconciler-type=lease \

roles/kube-master/templates/kube-controller-manager.service.j2

@@ -10,7 +10,7 @@ ExecStart={{ bin_dir }}/kube-controller-manager \
   --cluster-name=kubernetes \
   --cluster-signing-cert-file={{ ca_dir }}/ca.pem \
   --cluster-signing-key-file={{ ca_dir }}/ca-key.pem \
-  --kubeconfig=/etc/kubernetes/kube-controller-manager.kubeconfig \
+  --kubeconfig={{ cluster_dir }}/kube-controller-manager.kubeconfig \
   --leader-elect=true \
   --node-cidr-mask-size={{ NODE_CIDR_LEN }} \
   --root-ca-file={{ ca_dir }}/ca.pem \

roles/kube-master/templates/kube-scheduler.service.j2

@@ -5,7 +5,7 @@ Documentation=https://github.com/GoogleCloudPlatform/kubernetes
 [Service]
 ExecStart={{ bin_dir }}/kube-scheduler \
   --address=127.0.0.1 \
-  --kubeconfig=/etc/kubernetes/kube-scheduler.kubeconfig \
+  --kubeconfig={{ cluster_dir }}/kube-scheduler.kubeconfig \
   --leader-elect=true \
   --v=2
 Restart=always

roles/kube-master/vars/main.yml

@@ -0,0 +1,6 @@
+# etcd 集群服务地址列表, 根据etcd组成员自动生成
+TMP_ENDPOINTS: "{% for h in groups['etcd'] %}https://{{ h }}:2379,{% endfor %}"
+ETCD_ENDPOINTS: "{{ TMP_ENDPOINTS.rstrip(',') }}"
+
+# 设置 dns svc ip (这里选用 SERVICE_CIDR 中第1个IP)
+CLUSTER_KUBERNETES_SVC_IP: "{{ SERVICE_CIDR | ipaddr('net') | ipaddr(1) | ipaddr('address') }}"

roles/kube-node/defaults/main.yml

@@ -1,41 +0,0 @@
-# 设置 dns svc ip (这里选用 SERVICE_CIDR 中第2个IP)
-CLUSTER_DNS_SVC_IP: "{{ SERVICE_CIDR | ipaddr('net') | ipaddr(2) | ipaddr('address') }}"
-
-# 基础容器镜像
-SANDBOX_IMAGE: "easzlab/pause-amd64:3.2"
-
-# Kubelet 根目录
-KUBELET_ROOT_DIR: "/var/lib/kubelet"
-
-# node节点最大pod 数
-MAX_PODS: 110
-
-# 配置为kube组件（kubelet,kube-proxy,dockerd等）预留的资源量
-# 数值设置详见templates/kubelet-config.yaml.j2
-KUBE_RESERVED_ENABLED: "yes"
-
-# k8s 官方不建议草率开启 system-reserved, 除非你基于长期监控，了解系统的资源占用状况；
-# 并且随着系统运行时间，需要适当增加资源预留，数值设置详见templates/kubelet-config.yaml.j2
-# 系统预留设置基于 4c/8g 虚机，最小化安装系统服务，如果使用高性能物理机可以适当增加预留
-# 另外，集群安装时候apiserver等资源占用会短时较大，建议至少预留1g内存
-SYS_RESERVED_ENABLED: "no"
-
-# haproxy balance mode 
-BALANCE_ALG: "roundrobin"
-
-# 设置 APISERVER 地址
-KUBE_APISERVER: "{%- if inventory_hostname in groups['kube-master'] -%} \
-                     https://{{ inventory_hostname }}:6443 \
-                 {%- else -%} \
-                     {%- if groups['kube-master']|length > 1 -%} \
-                         https://127.0.0.1:6443 \
-                     {%- else -%} \
-                         https://{{ groups['kube-master'][0] }}:6443 \
-                     {%- endif -%} \
-                 {%- endif -%}"
-
-# 增加/删除 master 节点时，node 节点需要重新配置 haproxy 等
-MASTER_CHG: "no"
-
-# 离线安装 haproxy (offline|online)
-INSTALL_SOURCE: "online"

roles/kube-node/templates/kube-proxy.service.j2

@@ -10,7 +10,7 @@ ExecStart={{ bin_dir }}/kube-proxy \
   --bind-address={{ inventory_hostname }} \
   --cluster-cidr={{ CLUSTER_CIDR }} \
   --hostname-override={{ inventory_hostname }} \
-  --kubeconfig=/etc/kubernetes/kube-proxy.kubeconfig \
+  --kubeconfig={{ cluster_dir }}/kube-proxy.kubeconfig \
   --logtostderr=true \
   --proxy-mode={{ PROXY_MODE }}
 Restart=always

roles/kube-node/vars/main.yml

@@ -0,0 +1,13 @@
+# 设置 dns svc ip (这里选用 SERVICE_CIDR 中第2个IP)
+CLUSTER_DNS_SVC_IP: "{{ SERVICE_CIDR | ipaddr('net') | ipaddr(2) | ipaddr('address') }}"
+
+# 设置 APISERVER 地址
+KUBE_APISERVER: "{%- if inventory_hostname in groups['kube-master'] -%} \
+                     https://{{ inventory_hostname }}:6443 \
+                 {%- else -%} \
+                     {%- if groups['kube-master']|length > 1 -%} \
+                         https://127.0.0.1:6443 \
+                     {%- else -%} \
+                         https://{{ groups['kube-master'][0] }}:6443 \
+                     {%- endif -%} \
+                 {%- endif -%}"

roles/kube-ovn/vars/main.yml

@@ -1,10 +1,4 @@
-# 选择 OVN DB and OVN Control Plane 节点，默认为第一个master节点
-OVN_DB_NODE: "{{ groups['kube-master'][0] }}"
-
 kube_ovn_default_cidr: "{{ CLUSTER_CIDR }}"
 kube_ovn_default_gateway: "{{ CLUSTER_CIDR | ipaddr('net') | ipaddr(1) | ipaddr('address') }}"
 kube_ovn_node_switch_cidr: 100.64.0.0/16
 kube_ovn_enable_mirror: true
-
-# 离线镜像tar包
-kube_ovn_offline: "kube_ovn_0.9.1.tar"

roles/kube-router/defaults/main.yml

@@ -1,18 +0,0 @@
-# 更多设置，参考https://github.com/cloudnativelabs/kube-router/blob/master/docs/user-guide.md 
-
-# 因目前 kube-proxy 已提供 ipvs 模式，这里不使用 kube-router 的 service_proxy
-#SERVICE_PROXY: "false"
-
-# 公有云上存在限制，一般需要始终开启 ipinip；自有环境可以设置为 "subnet"
-OVERLAY_TYPE: "full"
-
-# NetworkPolicy 支持开关
-FIREWALL_ENABLE: "true"
-
-# kube-router 镜像版本
-kube_router_ver: "v0.3.1"
-busybox_ver: "1.28.4"
-
-# kube-router 离线镜像tar包
-kuberouter_offline: "kube-router_{{ kube_router_ver }}.tar"
-busybox_offline: "busybox_{{ busybox_ver }}.tar"

roles/prepare/defaults/main.yml

@@ -1,3 +0,0 @@
-# 离线安装系统软件包 (offline|online)
-INSTALL_SOURCE: "online"
-

roles/prepare/tasks/main.yml

@@ -38,7 +38,7 @@
 
 - block:
     - name: 分发证书相关
-      copy: src={{ base_dir }}/.cluster/ssl/{{ item }} dest={{ ca_dir }}/{{ item }}
+      copy: src={{ cluster_dir }}/ssl/{{ item }} dest={{ ca_dir }}/{{ item }}
       with_items:
       - admin.pem
       - admin-key.pem
@@ -57,13 +57,13 @@
       copy: src=/root/.kube/config dest=/root/.kube/config
 
     - name: 分发 kube-proxy.kubeconfig配置文件
-      copy: src={{ base_dir }}/.cluster/kube-proxy.kubeconfig dest=/etc/kubernetes/kube-proxy.kubeconfig
+      copy: src={{ cluster_dir }}/kube-proxy.kubeconfig dest={{ cluster_dir }}/kube-proxy.kubeconfig
 
     - name: 分发 kube-controller-manager.kubeconfig配置文件
-      copy: src={{ base_dir }}/.cluster/kube-controller-manager.kubeconfig dest=/etc/kubernetes/kube-controller-manager.kubeconfig
+      copy: src={{ cluster_dir }}/kube-controller-manager.kubeconfig dest={{ cluster_dir }}/kube-controller-manager.kubeconfig
       when: "inventory_hostname in groups['kube-master']"
 
     - name: 分发 kube-scheduler.kubeconfig配置文件
-      copy: src={{ base_dir }}/.cluster/kube-scheduler.kubeconfig dest=/etc/kubernetes/kube-scheduler.kubeconfig
+      copy: src={{ cluster_dir }}/kube-scheduler.kubeconfig dest={{ cluster_dir }}/kube-scheduler.kubeconfig
       when: "inventory_hostname in groups['kube-master']"
   when: "inventory_hostname in groups['kube-master'] or inventory_hostname in groups['kube-node']"

tools/basic-env-setup.sh

@@ -1,152 +0,0 @@
-#!/bin/bash
-
-# 本脚本提供如下功能，作者：Joey Yang, https://github.com/Code2Life
-# 1. 在Ubuntu/CentOS/Fedora/ArchLinux中自动化的安装python+ansible;
-# 2. clone kubeasz项目代码, 并将需要的二进制文件下载解压到/etc/ansible/bin中;
-# 另外, 相关的k8s二进制文件, 我同步到了个人在七牛上的CDN存储中（速度更快）, 方便大家下载: filecdn.code2life.top;
-# 
-# 使用方法：
-# 1. 支持带参数的运行, 如: ./basic-env-setup.sh k8s.193.tar.gz 指定不同的kubernetes二进制文件, 无参数时默认最新的k8s.1100.tar.gz (k8s 1.10.0 + etcd 3.3.2).
-# 2. 也可以在任何一台支持的linux设备运行：curl http://filecdn.code2life.top/kubeasz-basic-env-setup.sh | sh -s
-# 已经亲测centos7/ubuntu16.04/debian9/fedora27都是可以的, 二进制包下载速度贼快.脚本运行完毕之后, 只需到/etc/ansible目录下配置好hosts, 复制完ssh的公钥即可通过ansible-playbook迅速搭建集群了.
-
-set -e
-
-# curl http://filecdn.code2life.top/kubeasz-basic-env-setup.sh | sh -s
-
-# 默认1.10.0 版本的 Kubernetes
-bin_resource_url='http://filecdn.code2life.top/k8s.1100.tar.gz'
-
-# 如果参数指定k8s相关的bin以指定的为准, 例如: k8s.193.tar.gz
-if [ "$1" ];then
-  bin_resource_url="http://filecdn.code2life.top/"$1
-fi
-
-# 各Linux版本安装python/pip
-# ---------------------------
-
-# debian 默认的apt源在国内访问很慢, 可手动修改/etc/apt/sources.list修改为其他源
-# 以 debian 9 为例, source.list可修改为如下内容, ubuntu修改方法类似, 找到相应系统和版本的镜像源替换即可
-# deb http://mirrors.163.com/debian/  stretch main non-free contrib
-# deb http://mirrors.163.com/debian/  stretch-updates main non-free contrib
-# deb http://mirrors.163.com/debian/  stretch-backports main non-free contrib
-# deb http://mirrors.163.com/debian-security/  stretch/updates main non-free contrib
-basic_ubuntu_debian() {
-  echo "Setup Basic Environment for Ubuntu/Debian."
-  apt-get update && apt-get upgrade -y && apt-get dist-upgrade -y
-  apt-get install python2.7 git python-pip curl -y
-
-  if [ ! -f /usr/bin/python ];then
-    ln -s /usr/bin/python2.7 /usr/bin/python
-  fi
-}
-
-# 红帽系Liunx可修改yum源加快下载速度, 修改/etc/yum.repos.d内文件即可
-basic_centos() {
-  echo "Setup Basic Environment for CentOS."
-  yum install epel-release -y
-  yum update -y
-  yum erase firewalld firewalld-filesystem python-firewall -y
-  yum install git python python-pip curl -y
-}
-
-basic_fedora() {
-  echo "Setup Basic Environment for Fedora."
-  yum update -y
-  yum install git python python-pip curl -y
-}
-
-# archlinux 使用pacman进行包管理
-basic_arch() {
-  pacman -Syu --noconfirm
-  pacman -S python git python-pip curl --noconfirm
-}
-
-# 使用pip安装ansible, 并下载k8s相关bin文件
-setup_ansible_k8s() {
-  echo "Download Ansible and Kubernetes binaries."
-  pip install pip --upgrade -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
-  pip install --no-cache-dir ansible -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
-
-  git clone --depth=1 https://github.com/easzlab/kubeasz.git
-  mv kubeasz /etc/ansible
-
-  # Download from CDN & Move bin files
-  curl -o k8s_download.tar.gz "$bin_resource_url"
-  tar xvf k8s_download.tar.gz
-  mv -f bin/* /etc/ansible/bin
-  rm -rf bin
-  echo "Finish setup. Please config your hosts and run 'ansible-playbook' command at /etc/ansible."
-}
-# ---------------------------
-
-# 判断Linux发行版, 执行不同基础环境设置方法
-# ---------------------------
-lsb_dist=''
-command_exists() {
-    command -v "$@" > /dev/null 2>&1
-}
-if command_exists lsb_release; then
-    lsb_dist="$(lsb_release -si)"
-    lsb_version="$(lsb_release -rs)"
-fi
-if [ -z "$lsb_dist" ] && [ -r /etc/lsb-release ]; then
-    lsb_dist="$(. /etc/lsb-release && echo "$DISTRIB_ID")"
-    lsb_version="$(. /etc/lsb-release && echo "$DISTRIB_RELEASE")"
-fi
-if [ -z "$lsb_dist" ] && [ -r /etc/debian_version ]; then
-    lsb_dist='debian'
-fi
-if [ -z "$lsb_dist" ] && [ -r /etc/fedora-release ]; then
-    lsb_dist='fedora'
-fi
-if [ -z "$lsb_dist" ] && [ -r /etc/os-release ]; then
-    lsb_dist="$(. /etc/os-release && echo "$ID")"
-fi
-if [ -z "$lsb_dist" ] && [ -r /etc/centos-release ]; then
-    lsb_dist="$(cat /etc/*-release | head -n1 | cut -d " " -f1)"
-fi
-if [ -z "$lsb_dist" ] && [ -r /etc/redhat-release ]; then
-    lsb_dist="$(cat /etc/*-release | head -n1 | cut -d " " -f1)"
-fi
-lsb_dist="$(echo $lsb_dist | cut -d " " -f1)"
-lsb_dist="$(echo "$lsb_dist" | tr '[:upper:]' '[:lower:]')"
-# ---------------------------
-
-# ---------------------------
-setup_env(){
-    case "$lsb_dist" in
-        centos)
-        basic_centos
-        setup_ansible_k8s
-        exit 0
-    ;;
-        fedora)
-        basic_fedora
-        setup_ansible_k8s
-        exit 0
-    ;;
-        ubuntu)
-        basic_ubuntu_debian
-        setup_ansible_k8s
-        exit 0
-    ;;
-        debian)
-        basic_ubuntu_debian
-        setup_ansible_k8s
-        exit 0
-    ;;
-        arch)
-        basic_arch
-        setup_ansible_k8s
-        exit 0
-    ;;
-        suse)
-        echo 'Not implementation yet.'
-        exit 1
-    esac
-    echo "Error: Unsupported OS, please set ansible environment manually."
-    exit 1
-}
-setup_env
-# ---------------------------

tools/change_ip_aio.yml

@@ -1,43 +0,0 @@
-# 敬告：本脚本仅作测试交流使用，详细操作说明请参阅docs/op/change_ip_allinone.md
-# 此脚本仅用于allinone部署情况下，需要修改host ip地址使用（比如，虚机装完allinone部署，克隆或者复制分享给别人测试使用）
-#
-# ------- 前提 ：一个运行正常的allinone部署在虚机，关机后复制给别人使用，新虚机开机后如果需要修改IP，请执行如下步骤
-# 1. 修改ansible hosts文件：sed -i 's/$OLD_IP/$NEW_IP/g' /etc/ansible/hosts
-# 2. 配置ssh免密码登录：ssh-copy-id $NEW_IP 按提示完成
-# 3. 检查下修改是否成功，并且能够成功执行 ansible all -m ping
-# 4. 运行本脚本 ansible-playbook /etc/ansible/tools/change_ip_aio.yml
-
-- hosts: kube-master    # hosts 角色无所谓，反正allinone所有角色都是同个ip
-  tasks:
-  - name: 删除一些证书和配置，后面会以新IP重新生成
-    file: name={{ item }} state=absent
-    with_items:
-    - "/etc/etcd/ssl/etcd.pem"               # 删除etcd证书
-    - "/etc/kubernetes/ssl/kubernetes.pem"   # 删除旧master证书
-    - "/etc/kubernetes/kubelet.kubeconfig"   # 删除旧kubelet配置文件
-
-- hosts: kube-master
-  roles:
-  - deploy
-  - etcd
-  - kube-master
-  - kube-node
-
-- hosts: kube-master
-  tasks:
-  - name: 删除老IP地址的node
-    shell: "{{ bin_dir }}/kubectl get node |grep -v '{{ inventory_hostname }}'|awk '{print $1}' |xargs {{ bin_dir }}/kubectl delete node"
-    ignore_errors: true
-
-  - name: 删除原network插件部署
-    shell: "{{ bin_dir }}/kubectl delete -f /opt/kube/kube-system/{{ CLUSTER_NETWORK }}.yaml || \
-	    {{ bin_dir }}/kubectl delete -f /opt/kube/kube-ovn/"
-    ignore_errors: true
-
-- hosts: kube-master
-  roles:
-  - { role: calico, when: "CLUSTER_NETWORK == 'calico'" }
-  - { role: cilium, when: "CLUSTER_NETWORK == 'cilium'" }
-  - { role: flannel, when: "CLUSTER_NETWORK == 'flannel'" }
-  - { role: kube-router, when: "CLUSTER_NETWORK == 'kube-router'" }
-  - { role: kube-ovn, when: "CLUSTER_NETWORK == 'kube-ovn'" }

tools/change_k8s_network.yml

@@ -1,97 +0,0 @@
-# 重置k8s pod网络脚本，使用请仔细阅读 docs/op/change_k8s_network.md
-- hosts:
-  - kube-master
-  - kube-node
-  tasks:
-  - name: 获取所有已经创建的POD信息
-    command: "{{ bin_dir }}/kubectl get daemonset -n kube-system"
-    register: pod_info
-    run_once: true
-
-  - name: 删除原network插件部署
-    shell: "{{ bin_dir }}/kubectl delete -f /opt/kube/kube-system/{{ item }}.yaml"
-    with_items:
-    - calico
-    - cilium
-    - flannel
-    - kube-router
-    when: 'item in pod_info.stdout'
-    run_once: true
-    ignore_errors: true
-
-  - name: 清理kube-router相关
-    shell: "{{ bin_dir }}/docker run --privileged --net=host cloudnativelabs/kube-router --cleanup-config"
-    ignore_errors: true
-    when: '"kube-router" in pod_info.stdout'
-
-  - name: 停止 kube-node 相关服务
-    service: name={{ item }} state=stopped
-    with_items:
-    - kubelet
-    - kube-proxy
-    ignore_errors: true
-
-  - name: 清理calico残留路由
-    shell: "for rt in `ip route|grep bird|sed 's/blackhole//'|awk '{print $1}'`;do ip route del $rt;done;"
-    when: '"calico" in pod_info.stdout'
-    ignore_errors: true
-
-  - name: 清理 kube-proxy产生的iptables规则
-    shell: "{{ bin_dir }}/kube-proxy --cleanup"
-    ignore_errors: true
-
-  - name: 清理目录和文件
-    file: name={{ item }} state=absent
-    with_items:
-    - "/etc/cni/"
-    - "/run/flannel/"
-    - "/etc/calico/"
-    - "/var/run/calico/"
-    - "/var/lib/calico/"
-    - "/var/log/calico/"
-    - "/etc/cilium/"
-    - "/var/run/cilium/"
-    - "/sys/fs/bpf/tc/"
-    - "/var/lib/cni/"
-    - "/var/lib/kube-router/"
-    - "/opt/kube/kube-system/"
-
-  - name: 清理网络
-    shell: "ip link del tunl0; \
-        ip link del flannel.1; \
-        ip link del cni0; \
-        ip link del mynet0; \
-        ip link del kube-bridge; \
-        ip link del dummy0; \
-        ip link del kube-ipvs0; \
-        ip link del cilium_net; \
-        ip link del cilium_vxlan; \
-        systemctl restart networking; \
-        systemctl restart network"
-    ignore_errors: true
-
-  - name: 开启 kube-node 相关服务
-    service: name={{ item }} state=started enabled=yes
-    with_items:
-    - kubelet
-    - kube-proxy
-    ignore_errors: true
-
-- hosts:
-  - kube-master
-  - kube-node
-  # 安装新的网络插件
-  roles:
-  - { role: calico, when: "CLUSTER_NETWORK == 'calico'" }
-  - { role: cilium, when: "CLUSTER_NETWORK == 'cilium'" }
-  - { role: flannel, when: "CLUSTER_NETWORK == 'flannel'" }
-  - { role: kube-router, when: "CLUSTER_NETWORK == 'kube-router'" }
-
-- hosts: kube-node
-  tasks:
-  # 删除所有运行pod，由controller自动重建
-  - name: 重启所有pod
-    shell: "for NS in $({{ bin_dir }}/kubectl get ns|awk 'NR>1{print $1}'); \
-                do {{ bin_dir }}/kubectl delete pod --all -n $NS; done;"
-    ignore_errors: true
-    run_once: true