自动化部署K8S(v1.10.11)集群
自动化部署k8s集群，这里使用了腾讯的paas平台的标准运维和作业平台工具来辅助实现的。当然也可以不使用该平台，因为该平台依赖于gse-agent的服务。如果不依赖该平台，则跳过第一步，执行第二步即可。不过没有测试直接执行第二步脚本，应该需要按照自己的场景稍微修改一下，测试便可。后续流程和步骤都是一样的。以下介绍使用paas平台标准运维的实现。操作系统为：CentOS 7
标准运维实现自动化部署K8S集群主要分两步，第一步是部署gse-agent；第二步是部署k8s-master。
第一步：部署gse-agent。如下：
第二步：部署k8s集群。主要通过作业平台分为5小步执行对应的脚本实现。步骤如下：
按执行顺序列出对应的脚本，如下：
按执行顺序列出对应的脚本，如下：
1）master节点间ssh免秘钥登录设置 -- ssh_login.sh
#!/bin/bash
anynowtime="date +'%Y-%m-%d %H:%M:%S'"
NOW="echo [\`$anynowtime\`][PID:$$]"
##### 可在脚本开始运行时调用，打印当时的时间戳及PID。
function job_start
{
echo "`eval $NOW` job_start"
}
##### 可在脚本执行成功的逻辑分支处调用，打印当时的时间戳及PID。
function job_success
{
MSG="$*"
echo "`eval $NOW` job_success:[$MSG]"
exit 0
}
##### 可在脚本执行失败的逻辑分支处调用，打印当时的时间戳及PID。
function job_fail
{
MSG="$*"
echo "`eval $NOW` job_fail:[$MSG]"
exit 1
}
job_start
function set_ssh_login()
{
mkdir -p /root/.ssh/
echo "-----BEGIN RSA PRIVATE KEY-----
MIIEowIBAAKCAQEAs889Icr6U1aCTohdxQL+e/L5jIqmEBHzlTqTGb57b4vthxuT
n/GP3ccvBHwvdssdvsdvJwgWGZqHVeDOgzzkFbdl42wE2jW8A8lIhOS+xE
66w1VaO0Ii/+EwS7+FQHDyidYnlJ5dAPA/Uq+QIDAQABAoIBAD2BvHWcyzhKtVRL
zVehCJA5sydiHiANI/d+C+eYgvzLLrsysbLanM3OXsT3+IMrmWGLOYqAvCtsKCAx
fb1UW1N7uF4HO2FYrQsvdeVwew2FqMKNUQqvTlo+m3vKaxmeEOwg19aPXM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-----END RSA PRIVATE KEY-----" >> /root/.ssh/id_rsa
echo "ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQCzzz0hyvpTVoJOiF3FAv578vmMiqYQEfOVOpMZvntvi+2HG5Of8Y/dxy8EfB92SIrTxKeE+HiLgMhkPC3wLH9Y5oh1emi+y7/PV0OoXpf/A/pBr032BAFXMVOn5F8OgEuIQpN3J9o63zH0x7hIZGOGHNfp083Y1+/olbGBkggpyGeXEdwNtFh/acnCBYZmodV4M6DPOQVt2XjbATaNbwDyUiE5L7ETrrDVVo7QiL/4TBLv4VAcPKJ1ieUnl0A8D9Sr5 root@k8s" >> /root/.ssh/authorized_keys
chmod 700 /root/.ssh
chmod 600 /root/.ssh/id_rsa
chmod 600 /root/.ssh/authorized_keys
sed -i s/PermitRootLogin no/PermitRootLogin yes/g /etc/ssh/sshd_config
systemctl restart sshd
}
set_ssh_login
2) etcd集群部署 -- etcd_install.sh ${cluster_ip}
#!/bin/bash
anynowtime="date +'%Y-%m-%d %H:%M:%S'"
NOW="echo [\`$anynowtime\`][PID:$$]"
##### 可在脚本开始运行时调用，打印当时的时间戳及PID。
function job_start
{
echo "`eval $NOW` job_start"
}
##### 可在脚本执行成功的逻辑分支处调用，打印当时的时间戳及PID。
function job_success
{
MSG="$*"
echo "`eval $NOW` job_success:[$MSG]"
exit 0
}
##### 可在脚本执行失败的逻辑分支处调用，打印当时的时间戳及PID。
function job_fail
{
MSG="$*"
echo "`eval $NOW` job_fail:[$MSG]"
exit 1
}
job_start
master1=$1
master2=$2
master3=$3
systemctl stop firewalld
systemctl disable firewalld
setenforce 0
cat <<EOF > /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.ipv4.ip_forward=1
EOF
sysctl --system
iptables -P FORWARD ACCEPT
while true
do
tmp_ip=$(ip a | grep eth0 -C 0 | grep inet | awk '{print $2}' | awk -F "/" '{print $1}')
if [[ ${tmp_ip} == "" ]];then
ip=$(ip a | grep $(route -n | grep UG | awk '{print $NF}') -C 0 | grep inet | awk '{print $2}' | awk -F "/" '{print $1}')
else
ip=${tmp_ip}
fi
if [[ "x$ip" == "x" ]]; then
sleep 5
echo "waiting ip..."
else
break
fi
done
hostname=k8s-${ip//./-}
hostnamectl set-hostname --static $hostname
hostname1=k8s-${master1//./-}
hostname2=k8s-${master2//./-}
hostname3=k8s-${master3//./-}
echo "$master1 $hostname1" >> /etc/hosts
echo "$master2 $hostname2" >> /etc/hosts
echo "$master3 $hostname3" >> /etc/hosts
yum -y install etcd
if [[ $ip == $master1 ]]; then
etcd_name="etcd1"
elif [[ $ip == $master2 ]]; then
etcd_name="etcd2"
else
etcd_name="etcd3"
fi
cat << EOF > /etc/etcd/etcd.conf
ETCD_NAME=${etcd_name}
ETCD_DATA_DIR="/var/lib/etcd/default.etcd"
ETCD_LISTEN_PEER_URLS="http://0.0.0.0:2380"
ETCD_LISTEN_CLIENT_URLS="http://0.0.0.0:2379"
ETCD_INITIAL_ADVERTISE_PEER_URLS="http://${hostname}:2380"
ETCD_INITIAL_CLUSTER="etcd1=http://${hostname1}:2380,etcd2=http://${hostname2}:2380,etcd3=http://${hostname3}:2380"
ETCD_INITIAL_CLUSTER_STATE="new"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_ADVERTISE_CLIENT_URLS="http://${hostname}:2379"
EOF
cat << EOF > /usr/lib/systemd/system/etcd.service
[Unit]
Description=Etcd Server
After=network.target
After=network-online.target
Wants=network-online.target
[Service]
Type=notify
WorkingDirectory=/var/lib/etcd/
EnvironmentFile=-/etc/etcd/etcd.conf
User=etcd
# set GOMAXPROCS to number of processors
ExecStart=/bin/bash -c "GOMAXPROCS=$(nproc) /usr/bin/etcd --name="${ETCD_NAME}" --data-dir="${ETCD_DATA_DIR}" --listen-client-urls="${ETCD_LISTEN_CLIENT_URLS}""
Restart=on-failure
LimitNOFILE=65536
[Install]
WantedBy=multi-user.target
EOF
systemctl daemon-reload
systemctl enable etcd
systemctl start etcd
systemctl status etcd
sleep 10
etcdctl cluster-health
etcdctl member list
etcdctl mkdir /k8s
etcdctl set /k8s/network/config '{ "Network": "172.30.0.0/16", "SubnetLen":24,"Backend": { "Type": "vxlan" } }'
3)ssl证书生成 -- ssl_create.sh ${cluster_ip}
#!/bin/bash
anynowtime="date +'%Y-%m-%d %H:%M:%S'"
NOW="echo [\`$anynowtime\`][PID:$$]"
##### 可在脚本开始运行时调用，打印当时的时间戳及PID。
function job_start
{
echo "`eval $NOW` job_start"
}
##### 可在脚本执行成功的逻辑分支处调用，打印当时的时间戳及PID。
function job_success
{
MSG="$*"
echo "`eval $NOW` job_success:[$MSG]"
exit 0
}
##### 可在脚本执行失败的逻辑分支处调用，打印当时的时间戳及PID。
function job_fail
{
MSG="$*"
echo "`eval $NOW` job_fail:[$MSG]"
exit 1
}
job_start
mkdir -p /root/pki
mkdir -p /etc/kubernetes/pki
mkdir -p /etc/kubernetes/auth
while true
do
tmp_ip=$(ip a | grep eth0 -C 0 | grep inet | awk '{print $2}' | awk -F "/" '{print $1}')
if [[ ${tmp_ip} == "" ]];then
ip=$(ip a | grep $(route -n | grep UG | awk '{print $NF}') -C 0 | grep inet | awk '{print $2}' | awk -F "/" '{print $1}')
else
ip=${tmp_ip}
fi
if [[ "x$ip" == "x" ]]; then
sleep 5
echo "waiting ip..."
else
break
fi
done
if [[ $ip == $1 ]]; then
cd /root/pki
cat << EOF > csr.conf
[ req ]
default_bits = 2048
prompt = no
default_md = sha256
req_extensions = req_ext
distinguished_name = dn
[ dn ]
C = cn
ST = cn
L = cn
O = gz
OU = gz
CN = $1
[ req_ext ]
subjectAltName = IP:10.96.0.1
[ alt_names ]
DNS.1 = kubernetes
DNS.2 = kubernetes.default
DNS.3 = kubernetes.default.svc
DNS.4 = kubernetes.default.svc.cluster
IP.1 = $1
IP.2 = 10.96.0.1
[ v3_ext ]
authorityKeyIdentifier=keyid,issuer:always
basicConstraints=CA:FALSE
keyUsage=keyEncipherment,dataEncipherment
extendedKeyUsage=serverAuth,clientAuth
subjectAltName=@alt_names
EOF
openssl genrsa -out ca.key 2048
openssl req -x509 -new -nodes -key ca.key -config csr.conf -out ca.crt -days 10000
openssl genrsa -out server.key 2048
openssl req -new -key server.key -config csr.conf -out server.csr
openssl x509 -req -in server.csr -CA ca.crt -CAkey ca.key -CAcreateserial -days 10000 -extfile csr.conf -out server.crt
cp /root/pki/* /etc/kubernetes/pki/
chmod 600 /etc/kubernetes/pki/*
echo "*******,abc,1" > /etc/kubernetes/auth/auth.csv
echo "*******,abc,1" > /etc/kubernetes/auth/token.csv
else
scp -o stricthostkeychecking=no $1:/root/pki/* /etc/kubernetes/pki/
scp -o stricthostkeychecking=no $1:/etc/kubernetes/auth/* /etc/kubernetes/auth/
fi
4)k8s-master组件部署 -- k8s_master_install.sh ${cluster_ip} ${cluster_domain}
#!/bin/bash
anynowtime="date +'%Y-%m-%d %H:%M:%S'"
NOW="echo [\`$anynowtime\`][PID:$$]"
##### 可在脚本开始运行时调用，打印当时的时间戳及PID。
function job_start
{
echo "`eval $NOW` job_start"
}
##### 可在脚本执行成功的逻辑分支处调用，打印当时的时间戳及PID。
function job_success
{
MSG="$*"
echo "`eval $NOW` job_success:[$MSG]"
exit 0
}
##### 可在脚本执行失败的逻辑分支处调用，打印当时的时间戳及PID。
function job_fail
{
MSG="$*"
echo "`eval $NOW` job_fail:[$MSG]"
exit 1
}
job_start
master1=$1
master2=$2
master3=$3
cluster_domain=$4
while true
do
tmp_ip=$(ip a | grep eth0 -C 0 | grep inet | awk '{print $2}' | awk -F "/" '{print $1}')
if [[ ${tmp_ip} == "" ]];then
ip=$(ip a | grep $(route -n | grep UG | awk '{print $NF}') -C 0 | grep inet | awk '{print $2}' | awk -F "/" '{print $1}')
else
ip=${tmp_ip}
fi
if [[ "x$ip" == "x" ]]; then
sleep 5
echo "waiting ip..."
else
break
fi
done
function get_download_host_resolve()
{
iplist=("xxx.xxx.xxx.xxx" "xxx.xxx.xxx.xxx")
for ip in ${iplist[@]};do
ping -c 1 -W 3 $ip > /dev/null 2>&1
if [[ $? == 0 ]];then
echo $ip
break
fi
done
}
download_host_ip=$(get_download_host_resolve)
echo "use download_host_ip:$download_host_ip"
curl_option="-s --connect-timeout 3 --resolve doamin:80:$download_host_ip --retry 3"
index_list=`curl $curl_option $download_link`
[[ $? != 0 ]] && job_fail "donwload index_list fail"
echo "================= index list start ====================="
echo $index_list
echo "================= index list end ======================="
cd /root/
mkdir docker
mkdir kubernetes
for i in ${index_list[@]};do
echo "downloading... $i"
curl -O $curl_option $i
[[ $? != 0 ]] && job_fail "donwload file fail"
done
mv kube* /root/kubernetes/
cd /root/kubernetes
chmod 744 kube-controller-manager
chmod 744 kube-scheduler
chmod 744 kube-apiserver
chmod 744 kube-proxy
chmod 744 kubelet
chmod 744 kubectl
cp kube-controller-manager kube-scheduler kube-apiserver kubectl /usr/bin/
cp kube-controller-manager kube-scheduler kube-apiserver kubectl /bin/
cp kube-*.service /usr/lib/systemd/system/
systemctl daemon-reload
mkdir -p /var/log/kubernetes/kube-proxy
mkdir -p /var/log/kubernetes/kube-scheduler
mkdir -p /var/log/kubernetes/kubelet
mkdir -p /var/log/kubernetes/kube-controller-manager
mkdir -p /var/log/kubernetes/kube-apiserver
cat << EOF > /etc/kubernetes/kube-apiserver
KUBE_OTHER_OPTIONS="--kubelet-preferred-address-types=InternalIP"
KUBE_LOGTOSTDERR="--logtostderr=false"
KUBE_LOG_DIR="--log-dir=/var/log/kubernetes/kube-apiserver"
KUBE_LOG_LEVEL="--v=5"
KUBE_ETCD_SERVERS="--etcd-servers=http://${master1}:2379,http://${master2}:2379,http://${master3}:2379"
KUBE_API_ADDRESS="--insecure-bind-address=0.0.0.0"
KUBE_API_PORT="--secure-port=6443"
KUBE_ADVERTISE_ADDR="--advertise-address=${ip}"
KUBE_ALLOW_PRIV="--allow-privileged=true"
KUBE_SERVICE_ADDRESSES="--service-cluster-ip-range=10.96.0.0/16"
KUBE_ADMISSION_CONTROL="--admission-control=Initializers,NamespaceLifecycle,ServiceAccount,DefaultTolerationSeconds,ResourceQuota"
KUBE_API_TLS_CERT_FILE="--tls-cert-file=/etc/kubernetes/pki/server.crt"
KUBE_API_TLS_PRIVATE_KEY_FILE="--tls-private-key-file=/etc/kubernetes/pki/server.key"
KUBE_API_TOKEN_AUTH_ARGS="--token-auth-file=/etc/kubernetes/auth/token.csv"
KUBE_API_BASIC_AUTH_ARGS="--basic-auth-file=/etc/kubernetes/auth/auth.csv"
KUBE_NEW_FEATURES="--feature-gates=CustomPodDNS=true"
KUBE_APISERVER_COUNT="--apiserver-count=3"
EOF
cat << EOF > /etc/kubernetes/kube-controller-manager
KUBE_MASTER="--master=https://${cluster_domain}"
KUBE_LOGTOSTDERR="--logtostderr=false"
KUBE_LOG_LEVEL="--v=5"
KUBE_LOG_DIR="--log-dir=/var/log/kubernetes/kube-controller-manager"
KUBE_CONFIG="--kubeconfig=/etc/kubernetes/kubelet.kubeconfig"
KUBE_CONTROLLER_MANAGER_ROOT_CA_FILE="--root-ca-file=/etc/kubernetes/pki/ca.crt"
KUBE_CONTROLLER_MANAGER_SERVICE_ACCOUNT_PRIVATE_KEY_FILE="--service-account-private-key-file=/etc/kubernetes/pki/server.key"
KUBE_LEADER_ELECT="--leader-elect=true"
KUBE_CONTROLLER_OTHER_OPTIONS="--terminated-pod-gc-threshold=1"
KUBE_NEW_FEATURES="--feature-gates=CustomPodDNS=true"
EOF
cat << EOF > /etc/kubernetes/kube-scheduler
KUBE_LOGTOSTDERR="--logtostderr=false"
KUBE_MASTER="--master=https://${cluster_domain}"
KUBE_LOG_LEVEL="--v=5"
KUBE_LOG_DIR="--log-dir=/var/log/kubernetes/kube-scheduler"
KUBE_CONFIG="--kubeconfig=/etc/kubernetes/kubelet.kubeconfig"
KUBE_LEADER_ELECT="--leader-elect=true"
KUBE_SCHEDULER_ARGS=""
KUBE_NEW_FEATURES="--feature-gates=CustomPodDNS=true"
EOF
cat << EOF > /etc/kubernetes/kubelet.kubeconfig
apiVersion: v1
kind: Config
clusters:
- cluster:
server: https://${cluster_domain}
insecure-skip-tls-verify: true
name: master-server
contexts:
- context:
cluster: master-server
user: remote-api-auth
name: default
current-context: default
users:
- name: remote-api-auth
user:
password: *******
username: abc
EOF
systemctl enable kube-apiserver kube-controller-manager kube-scheduler
systemctl start kube-apiserver kube-controller-manager kube-scheduler
systemctl status kube-apiserver kube-controller-manager kube-scheduler
sleep 10
5)k8s-minion节点部署 -- start_master_and_install_minion.sh ${cluster_ip} ${cluster_domain}
#!/bin/bash
anynowtime="date +'%Y-%m-%d %H:%M:%S'"
NOW="echo [\`$anynowtime\`][PID:$$]"
##### 可在脚本开始运行时调用，打印当时的时间戳及PID。
function job_start
{
echo "`eval $NOW` job_start"
}
##### 可在脚本执行成功的逻辑分支处调用，打印当时的时间戳及PID。
function job_success
{
MSG="$*"
echo "`eval $NOW` job_success:[$MSG]"
exit 0
}
##### 可在脚本执行失败的逻辑分支处调用，打印当时的时间戳及PID。
function job_fail
{
MSG="$*"
echo "`eval $NOW` job_fail:[$MSG]"
exit 1
}
job_start
master1=$1
master2=$2
master3=$3
cluster_domain=$4
cd /root/
tar -zxvf master_install_scripts.tar.gz
cat << EOF > install_node.sh
#!/bin/bash
K8S_API_URL="https://${cluster_domain}"
ETCD_URL="http://${master1}:2379,http://${master2}:2379,http://${master3}:2379"
POD_BASE_IMAGE="harbor-domain/library/pod-infrastructure:latest"
yum install -y wget telnet net-tools
bash +x *.sh
./1.yum.sh
./2.flannel.sh "${ETCD_URL}"
./3.docker.sh
./4.install-kube-proxy.sh
./5.kubelet.sh "${K8S_API_URL}" "${POD_BASE_IMAGE}"
./6.config-kube-proxy.sh
./7.firewalld.sh
./8.restart-service.sh
./9.nfs.sh
EOF
chmod +x install_node.sh
bash install_node.sh
bash register_k8s_inner.sh
bash 8.restart-service.sh
docker version
说明：在部署完成master集群后，如果需要加入minion节点，则只需要执行5）步中的脚本即可。
v1.10.11版本下载地址：https://storage.googleapis.com/kubernetes-release/release/v1.10.11/kubernetes-server-linux-amd64.tar.gz
下载后解压在kubernetes/server/bin目录下获取：kube-apiserver、kube-controller-manager、kube-scheduler、kube-proxy、kubelet、kubectl。
其他依赖的脚本如下：
#cat kube-apiserver.service
[Unit]
Description=Kubernetes API Server
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=-/etc/kubernetes/kube-apiserver
ExecStart=/usr/bin/kube-apiserver ${KUBE_LOGTOSTDERR}
${KUBE_LOG_LEVEL}
${KUBE_LOG_DIR}
${KUBE_OTHER_OPTIONS}
${KUBE_ETCD_SERVERS}
${KUBE_ETCD_CAFILE}
${KUBE_ETCD_CERTFILE}
${KUBE_ETCD_KEYFILE}
${KUBE_API_ADDRESS}
${KUBE_API_PORT}
${NODE_PORT}
${KUBE_ADVERTISE_ADDR}
${KUBE_ALLOW_PRIV}
${KUBE_SERVICE_ADDRESSES}
${KUBE_ADMISSION_CONTROL}
${KUBE_API_CLIENT_CA_FILE}
${KUBE_API_TLS_CERT_FILE}
${KUBE_API_TLS_PRIVATE_KEY_FILE}
${KUBE_API_TOKEN_AUTH_ARGS}
${KUBE_NEW_FEATURES}
${KUBE_API_BASIC_AUTH_ARGS}
${KUBE_APISERVER_COUNT}
Restart=on-failure
LimitCORE=infinity
LimitNOFILE=65536
LimitNPROC=65536
[Install]
WantedBy=multi-user.target
#cat kube-controller-manager.service
[Unit]
Description=Kubernetes Controller Manager
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=-/etc/kubernetes/kube-controller-manager
ExecStart=/usr/bin/kube-controller-manager ${KUBE_LOGTOSTDERR}
${KUBE_LOG_LEVEL}
${KUBE_CONTROLLER_OTHER_OPTIONS}
${KUBE_LOG_DIR}
${KUBE_MASTER}
${KUBE_CONFIG}
${KUBE_CONTROLLER_MANAGER_ROOT_CA_FILE}
${KUBE_CONTROLLER_MANAGER_SERVICE_ACCOUNT_PRIVATE_KEY_FILE}
${KUBE_NEW_FEATURES}
${KUBE_LEADER_ELECT}
Restart=on-failure
LimitCORE=infinity
LimitNOFILE=65536
LimitNPROC=65536
[Install]
WantedBy=multi-user.target
#cat kube-scheduler.service
[Unit]
Description=Kubernetes Scheduler
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=-/etc/kubernetes/kube-scheduler
ExecStart=/usr/bin/kube-scheduler ${KUBE_LOGTOSTDERR}
${KUBE_LOG_LEVEL}
${KUBE_LOG_DIR}
${KUBE_MASTER}
${KUBE_CONFIG}
${KUBE_LEADER_ELECT}
${KUBE_NEW_FEATURES}
$KUBE_SCHEDULER_ARGS
Restart=on-failure
LimitCORE=infinity
LimitNOFILE=65536
LimitNPROC=65536
[Install]
WantedBy=multi-user.target
经过测试，部署好master节点只需3-5分钟左右即可，效率极高。
测试集群：
# kubectl create namespace huangjie
# cat app_nginx.yaml
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: app-nginx
namespace: huangjie
spec:
replicas: 5
template:
metadata:
labels:
app: app-nginx
spec:
containers:
- name: app-nginx
image: harhor-domain/hj-demo/nginx:latest
ports:
- containerPort: 80
lifecycle:
preStop:
exec:
command: ["sleep","5"]
# cat app_nginx-svc.yaml
apiVersion: v1
kind: Service
metadata:
name: app-nginx
namespace: huangjie
spec:
clusterIP: None
ports:
- name: web
port: 8090
protocol: TCP
targetPort: 80
selector:
app: app-nginx
sessionAffinity: None
type: ClusterIP
status:
loadBalancer: {}
# kubectl apply -f app_nginx.yaml
# kubectl apply -f app_nginx-svc.yaml
状态：
测试集群内访问：
