etcd管理，证书配置，扩展，迁移恢复，带证书扩展节点

Posted 2020-12-29

广告 | kubernetes各版本离线安装包

etcd 证书配置

生产环境中给etcd配置证书相当重要，如果没有证书，那么k8s集群很容易被×××利用而去挖矿什么的。做法非常简单，比如你下了一个不安全的镜像，通过程序扫描到etcd的ip和端口，那么×××就可以绕开apiserver的认证直接写数据，写一些deployment pod等等，apiserver就会读到这些，从而去部署×××的程序。 我们就有一个集群这样被利用去挖矿了,安全无小事，如果×××恶意×××也可轻松删除你的所有数据，所以证书与定期备份都很重要,即便有多个etcd节点，本文深入探讨etcd管理的重要的几个东西。

证书生成

cfssl安装：

mkdir ~/bin
curl -s -L -o ~/bin/cfssl https://pkg.cfssl.org/R1.2/cfssl_linux-amd64
curl -s -L -o ~/bin/cfssljson https://pkg.cfssl.org/R1.2/cfssljson_linux-amd64
chmod +x ~/bin/{cfssl,cfssljson}
export PATH=$PATH:~/bin

mkdir ~/cfssl
cd ~/cfssl

写入如下json文件，ip替换成自己的

[email protected] cfssl]# cat ca-config.json
{
    "signing": {
        "default": {
            "expiry": "43800h"
        },
        "profiles": {
            "server": {
                "expiry": "43800h",
                "usages": [
                    "signing",
                    "key encipherment",
                    "server auth"
                ]
            },
            "client": {
                "expiry": "43800h",
                "usages": [
                    "signing",
                    "key encipherment",
                    "client auth"
                ]
            },
            "peer": {
                "expiry": "43800h",
                "usages": [
                    "signing",
                    "key encipherment",
                    "server auth",
                    "client auth"
                ]
            }
        }
    }
}
[[email protected] cfssl]# cat ca-csr.json
{
    "CN": "My own CA",
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
        {
            "C": "US",
            "L": "CA",
            "O": "My Company Name",
            "ST": "San Francisco",
            "OU": "Org Unit 1",
            "OU": "Org Unit 2"
        }
    ]
}
[[email protected] cfssl]# cat server.json
{
    "CN": "etcd0",
    "hosts": [
        "127.0.0.1",
        "0.0.0.0",
        "10.1.86.201",
        "10.1.86.203",
        "10.1.86.202"
    ],
    "key": {
        "algo": "ecdsa",
        "size": 256
    },
    "names": [
        {
            "C": "US",
            "L": "CA",
            "ST": "San Francisco"
        }
    ]
}

[[email protected] cfssl]# cat member1.json  # 填本机IP
{
    "CN": "etcd0",
    "hosts": [
        "10.1.86.201"
    ],
    "key": {
        "algo": "ecdsa",
        "size": 256
    },
    "names": [
        {
            "C": "US",
            "L": "CA",
            "ST": "San Francisco"
        }
    ]
}

[[email protected] cfssl]# cat client.json
{
    "CN": "client",
    "hosts": [
       ""
    ],
    "key": {
        "algo": "ecdsa",
        "size": 256
    },
    "names": [
        {
            "C": "US",
            "L": "CA",
            "ST": "San Francisco"
        }
    ]
}

生成证书：

cfssl gencert -initca ca-csr.json | cfssljson -bare ca -
cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=server server.json | cfssljson -bare server
cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=peer member1.json | cfssljson -bare member1
cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=client client.json | cfssljson -bare client

启动etcd

cfssl目录拷贝到/etc/kubernetes/pki/cfssl 目录

[[email protected] manifests]# cat etcd.yaml
apiVersion: v1
kind: Pod
metadata:
  annotations:
    scheduler.alpha.kubernetes.io/critical-pod: ""
  creationTimestamp: null
  labels:
    component: etcd
    tier: control-plane
  name: etcd
  namespace: kube-system
spec:
  containers:
  - command:
    - etcd
    - --advertise-client-urls=https://10.1.86.201:2379
    - --cert-file=/etc/kubernetes/pki/etcd/server.pem
    - --client-cert-auth=true
    - --data-dir=/var/lib/etcd
    - --initial-advertise-peer-urls=https://10.1.86.201:2380
    - --initial-cluster=etcd0=https://10.1.86.201:2380
    - --key-file=/etc/kubernetes/pki/etcd/server-key.pem
    - --listen-client-urls=https://10.1.86.201:2379
    - --listen-peer-urls=https://10.1.86.201:2380
    - --name=etcd0
    - --peer-cert-file=/etc/kubernetes/pki/etcd/member1.pem
    - --peer-client-cert-auth=true
    - --peer-key-file=/etc/kubernetes/pki/etcd/member1-key.pem
    - --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    - --snapshot-count=10000
    - --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    image: k8s.gcr.io/etcd-amd64:3.2.18
    imagePullPolicy: IfNotPresent
   #livenessProbe:
   #  exec:
   #    command:
   #    - /bin/sh
   #    - -ec
   #    - ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.201]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.pem
   #      --cert=/etc/kubernetes/pki/etcd/client.pem --key=/etc/kubernetes/pki/etcd/client-key.pem
   #      get foo
   #  failureThreshold: 8
   #  initialDelaySeconds: 15
   #  timeoutSeconds: 15
    name: etcd
    resources: {}
    volumeMounts:
    - mountPath: /var/lib/etcd
      name: etcd-data
    - mountPath: /etc/kubernetes/pki/etcd
      name: etcd-certs
  hostNetwork: true
  priorityClassName: system-cluster-critical
  volumes:
  - hostPath:
      path: /var/lib/etcd
      type: DirectoryOrCreate
    name: etcd-data
  - hostPath:
      path: /etc/kubernetes/pki/cfssl
      type: DirectoryOrCreate
    name: etcd-certs
status: {}

进入etcd容器执行：

alias etcdv3="ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.201]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.pem --cert=/etc/kubernetes/pki/etcd/client.pem --key=/etc/kubernetes/pki/etcd/client-key.pem"
etcdv3 member add etcd1 --peer-urls="https://10.1.86.202:2380"

增加节点

拷贝etcd0(10.1.86.201)节点上的证书到etcd1(10.1.86.202)节点上
修改member1.json:

{
    "CN": "etcd1",
    "hosts": [
        "10.1.86.202"
    ],
    "key": {
        "algo": "ecdsa",
        "size": 256
    },
    "names": [
        {
            "C": "US",
            "L": "CA",
            "ST": "San Francisco"
        }
    ]
}

重新生成在etcd1上生成member1证书：

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=peer member1.json | cfssljson -bare member1

启动etcd1：

[[email protected] manifests]# cat etcd.yaml
apiVersion: v1
kind: Pod
metadata:
  annotations:
    scheduler.alpha.kubernetes.io/critical-pod: ""
  creationTimestamp: null
  labels:
    component: etcd
    tier: control-plane
  name: etcd
  namespace: kube-system
spec:
  containers:
  - command:
    - etcd
    - --advertise-client-urls=https://10.1.86.202:2379
    - --cert-file=/etc/kubernetes/pki/etcd/server.pem
    - --data-dir=/var/lib/etcd
    - --initial-advertise-peer-urls=https://10.1.86.202:2380
    - --initial-cluster=etcd0=https://10.1.86.201:2380,etcd1=https://10.1.86.202:2380
    - --key-file=/etc/kubernetes/pki/etcd/server-key.pem
    - --listen-client-urls=https://10.1.86.202:2379
    - --listen-peer-urls=https://10.1.86.202:2380
    - --name=etcd1
    - --peer-cert-file=/etc/kubernetes/pki/etcd/member1.pem
    - --peer-client-cert-auth=true
    - --peer-key-file=/etc/kubernetes/pki/etcd/member1-key.pem
    - --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    - --snapshot-count=10000
    - --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    - --initial-cluster-state=existing  # 千万别加双引号，被坑死
    image: k8s.gcr.io/etcd-amd64:3.2.18
    imagePullPolicy: IfNotPresent
  # livenessProbe:
  #   exec:
  #     command:
  #     - /bin/sh
  #     - -ec
  #     - ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.202]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt
  #       --cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key
  #       get foo
  #   failureThreshold: 8
  #   initialDelaySeconds: 15
  #   timeoutSeconds: 15
    name: etcd
    resources: {}
    volumeMounts:
    - mountPath: /var/lib/etcd
      name: etcd-data
    - mountPath: /etc/kubernetes/pki/etcd
      name: etcd-certs
  hostNetwork: true
  priorityClassName: system-cluster-critical
  volumes:
  - hostPath:
      path: /var/lib/etcd
      type: DirectoryOrCreate
    name: etcd-data
  - hostPath:
      path: /etc/kubernetes/pki/cfssl
      type: DirectoryOrCreate
    name: etcd-certs
status: {}

或者用docker起先测试一下：

docker run --net=host -v /etc/kubernetes/pki/cfssl:/etc/kubernetes/pki/etcd k8s.gcr.io/etcd-amd64:3.2.18 etcd --advertise-client-urls=https://10.1.86.202:2379 --cert-file=/etc/kubernetes/pki/etcd/server.pem --data-dir=/var/lib/etcd --initial-advertise-peer-urls=https://10.1.86.202:2380 --initial-cluster=etcd0=https://10.1.86.201:2380,etcd1=https://10.1.86.202:2380 --key-file=/etc/kubernetes/pki/etcd/server-key.pem  --listen-client-urls=https://10.1.86.202:2379 --listen-peer-urls=https://10.1.86.202:2380 --name=etcd1 --peer-cert-file=/etc/kubernetes/pki/etcd/member1.pem --peer-key-file=/etc/kubernetes/pki/etcd/member1-key.pem --peer-client-cert-auth=true --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem --snapshot-count=10000 --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem --initial-cluster-state="existing"

etcd0上检查集群健康:

# etcdctl --endpoints=https://[10.1.86.201]:2379 --ca-file=/etc/kubernetes/pki/etcd/ca.pem --cert-file=/etc/kubernetes/pki/etcd/client.pem --key-file=/etc/kubernetes/pki/etcd/client-key.pem cluster-heal
th
member 5856099674401300 is healthy: got healthy result from https://10.1.86.201:2379
member df99f445ac908d15 is healthy: got healthy result from https://10.1.86.202:2379
cluster is healthy

etcd2增加同理，略

apiserver etcd证书 配置：

- --etcd-cafile=/etc/kubernetes/pki/cfssl/ca.pem
- --etcd-certfile=/etc/kubernetes/pki/cfssl/client.pem
- --etcd-keyfile=/etc/kubernetes/pki/cfssl/client-key.pem

快照与扩展节点

etcd快照恢复

说明：
有证书集群以下所有命令需带上如下证书参数，否则访问不了

--cacert=/etc/kubernetes/pki/etcd/ca.crt --cert=/etc/kubernetes/pki/etcd/peer.crt --key=/etc/kubernetes/pki/etcd/peer.key

endpoints默认为127.0.0.1:2379，若需指定远程etcd地址，可通过如下参数指定

--endpoints 172.16.154.81:2379

1、获取数据快照

ETCDCTL_API=3 etcdctl snapshot save snapshot.db

2、从快照恢复数据

ETCDCTL_API=3 etcdctl snapshot restore snapshot.db --data-dir=/var/lib/etcd/

3、启动新etcd节点，指定--data-dir=/var/lib/etcd/

etcd节点扩展

节点名	IP	备注
infra0	172.16.154.81	初始节点，k8s的master节点，kubeadm所部署的单节点etcd所在机器
infra1	172.16.154.82	待添加节点，k8s的node节点
infra2	172.16.154.83	待添加节点，k8s的node节点

1、从初始etcd节点获取数据快照

ETCDCTL_API=3 etcdctl --cacert=/etc/kubernetes/pki/etcd/ca.crt --cert=/etc/kubernetes/pki/etcd/peer.crt --key=/etc/kubernetes/pki/etcd/peer.key --endpoints=https://127.0.0.1:2379 snapshot save snapshot.db

2、将快照文件snapshot.db复制到infra1节点，并执行数据恢复命令

数据恢复命令

ETCDCTL_API=3 etcdctl snapshot restore snapshot.db --data-dir=/var/lib/etcd/

注：执行上述命令需要机器上有etcdctl

上述命令执行成功会将快照中的数据存放到/var/lib/etcd目录中

3、在infra1节点启动etcd
将如下yaml放入/etc/kubernetes/manifests

apiVersion: v1
kind: Pod
metadata:
  labels:
    component: etcd
    tier: control-plane
  name: etcd-172.16.154.82
  namespace: kube-system
spec:
  containers:
  - command:
    - etcd
    - --name=infra0
    - --initial-advertise-peer-urls=http://172.16.154.82:2380
    - --listen-peer-urls=http://172.16.154.82:2380
    - --listen-client-urls=http://172.16.154.82:2379,http://127.0.0.1:2379
    - --advertise-client-urls=http://172.16.154.82:2379
    - --data-dir=/var/lib/etcd
    - --initial-cluster-token=etcd-cluster-1
    - --initial-cluster=infra0=http://172.16.154.82:2380
    - --initial-cluster-state=new
    image: hub.xfyun.cn/k8s/etcd-amd64:3.1.12
    livenessProbe:
      httpGet:
        host: 127.0.0.1
        path: /health
        port: 2379
        scheme: HTTP
      failureThreshold: 8
      initialDelaySeconds: 15
      timeoutSeconds: 15
    name: etcd
    volumeMounts:
    - name: etcd-data
      mountPath: /var/lib/etcd
  hostNetwork: true
  volumes:
  - hostPath:
      path: /var/lib/etcd
      type: DirectoryOrCreate
    name: etcd-data

4、infra2节点加入etcd集群中
在infra1中etcd容器中执行

ETCDCTL_API=3 etcdctl member add infra2 --peer-urls="http://172.16.154.83:2380"

将如下yaml放入/etc/kubernetes/manifests，由kubelet启动etcd容器

apiVersion: v1
kind: Pod
metadata:
  labels:
    component: etcd
    tier: control-plane
  name: etcd-172.16.154.83
  namespace: kube-system
spec:
  containers:
  - command:
    - etcd
    - --name=infra1
    - --initial-advertise-peer-urls=http://172.16.154.83:2380
    - --listen-peer-urls=http://172.16.154.83:2380
    - --listen-client-urls=http://172.16.154.83:2379,http://127.0.0.1:2379
    - --advertise-client-urls=http://172.16.154.83:2379
    - --data-dir=/var/lib/etcd
    - --initial-cluster-token=etcd-cluster-1
    - --initial-cluster=infra1=http://172.16.154.82:2380,infra2=http://172.16.154.83:2380
    - --initial-cluster-state=existing
    image: hub.xfyun.cn/k8s/etcd-amd64:3.1.12
    livenessProbe:
      httpGet:
        host: 127.0.0.1
        path: /health
        port: 2379
        scheme: HTTP
      failureThreshold: 8
      initialDelaySeconds: 15
      timeoutSeconds: 15
    name: etcd
    volumeMounts:
    - name: etcd-data
      mountPath: /var/lib/etcd
  hostNetwork: true
  volumes:
  - hostPath:
      path: /home/etcd
      type: DirectoryOrCreate
    name: etcd-data

infra0节点加入集群重复上述操作；注意在加入集群之前，将之前/var/lib/etcd/的数据删除。

实践 - 给kubeadm单etcd增加etcd节点

环境介绍

10.1.86.201 单点etcd etcd0

10.1.86.202 扩展节点 etcd1

10.1.86.203 扩展节点 etcd2

安装k8s

先在etcd0节点上启动k8s，当然是使用sealyun的安装包 三步安装不多说

×××

按照上述生成证书的方法生成证书并拷贝到对应目录下

cp -r cfssl/ /etc/kubernetes/pki/

修改etcd配置：

cd /etc/kubernetes/manifests/
mv etcd.yaml ..   # 不要直接修改，防止k8s去读swap文件
vim ../etcd.yaml

vim里面全局替换，把127.0.0.1替换成ip地址

:%s/127.0.0.1/10.1.86.201/g

注释掉健康检测探针，否则加节点时健康检测会导致etcd0跪掉

#   livenessProbe:
#     exec:
#       command:
#       - /bin/sh
#       - -ec
#       - ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.201]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt
#         --cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key
#         get foo
#     failureThreshold: 8
#     initialDelaySeconds: 15
#     timeoutSeconds: 15

×××挂载配置目录

  volumes:
  - hostPath:
      path: /etc/kubernetes/pki/cfssl
      type: DirectoryOrCreate
    name: etcd-certs

×××配置,全改完长这样：

[[email protected] manifests]# cat ../etcd.yaml
apiVersion: v1
kind: Pod
metadata:
  annotations:
    scheduler.alpha.kubernetes.io/critical-pod: ""
  creationTimestamp: null
  labels:
    component: etcd
    tier: control-plane
  name: etcd
  namespace: kube-system
spec:
  containers:
  - command:
    - etcd
    - --advertise-client-urls=https://10.1.86.201:2379
    - --cert-file=/etc/kubernetes/pki/etcd/server.pem
    - --client-cert-auth=true
    - --data-dir=/var/lib/etcd
    - --initial-advertise-peer-urls=https://10.1.86.201:2380
    - --initial-cluster=etcd0=https://10.1.86.201:2380
    - --key-file=/etc/kubernetes/pki/etcd/server-key.pem
    - --listen-client-urls=https://10.1.86.201:2379
    - --listen-peer-urls=https://10.1.86.201:2380
    - --name=dev-86-201
    - --peer-cert-file=/etc/kubernetes/pki/etcd/member1.pem
    - --peer-client-cert-auth=true
    - --peer-key-file=/etc/kubernetes/pki/etcd/member1-key.pem
    - --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    - --snapshot-count=10000
    - --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    image: k8s.gcr.io/etcd-amd64:3.2.18
    imagePullPolicy: IfNotPresent
#   livenessProbe:
#     exec:
#       command:
#       - /bin/sh
#       - -ec
#       - ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.201]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt
#         --cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key
#         get foo
#     failureThreshold: 8
#     initialDelaySeconds: 15
#     timeoutSeconds: 15
    name: etcd
    resources: {}
    volumeMounts:
    - mountPath: /var/lib/etcd
      name: etcd-data
    - mountPath: /etc/kubernetes/pki/etcd
      name: etcd-certs
  hostNetwork: true
  priorityClassName: system-cluster-critical
  volumes:
  - hostPath:
      path: /etc/kubernetes/pki/cfssl
      type: DirectoryOrCreate
    name: etcd-certs
  - hostPath:
      path: /var/lib/etcd
      type: DirectoryOrCreate
    name: etcd-data
status: {}

启动etcd, 把yaml文件移回来：

mv ../etcd.yaml .

修改APIserver参数：

mv kube-apiserver.yaml ..
vim ../kube-apiserver.yaml

    - --etcd-cafile=/etc/kubernetes/pki/cfssl/ca.pem
    - --etcd-certfile=/etc/kubernetes/pki/cfssl/client.pem
    - --etcd-keyfile=/etc/kubernetes/pki/cfssl/client-key.pem
    - --etcd-servers=https://10.1.86.201:2379

启动apiserver:

mv ../kube-apiserver.yaml .

验证：

kubectl get pod -n kube-system  # 能正常返回pod标志成功

到此etcd0上的操作完成

增加新节点, 进入到etcd容器内:

[[email protected] ~]# docker exec -it a7001397e1e5 sh
/ # alias etcdv3="ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.201]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.pem --cert=/etc/kubernetes/pki/etcd/client.pem --key=/etc/kubernetes/pki/etcd/client-key
.pem"
/ # etcdv3 member update a874c87fd42044f  --peer-urls="https://10.1.86.201:2380" # 更新peer url 很重要
/ # etcdv3 member add etcd1 --peer-urls="https://10.1.86.202:2380"
Member 20c2a99381581958 added to cluster c9be114fc2da2776

ETCD_NAME="etcd1"
ETCD_INITIAL_CLUSTER="dev-86-201=https://127.0.0.1:2380,etcd1=https://10.1.86.202:2380"
ETCD_INITIAL_CLUSTER_STATE="existing"

/ # alias etcdv2="ETCDCTL_API=2 etcdctl --endpoints=https://[10.1.86.201]:2379 --ca-file=/etc/kubernetes/pki/etcd/ca.pem --cert-file=/etc/kubernetes/pki/etcd/client.pem --key-file=/etc/kubernetes/pki/etcd/client-key.pem"
/ # etcdv2 cluster-health

etcd1上增加一个etcd节点

同样先在etcd1（10.1.86.202) 上安装k8s，同etcd0上的安装

把etcd0的cfssl证书目录拷贝到etcd1上备用

scp -r [email protected]:/etc/kubernetes/pki/cfssl /etc/kubernetes/pki

修改member1.json:

[[email protected] cfssl]# cat member1.json
{
    "CN": "etcd1",      # CN 改一下
    "hosts": [
        "10.1.86.202"   # 主要改成自身ip
    ],
    "key": {
        "algo": "ecdsa",
        "size": 256
    },
    "names": [
        {
            "C": "US",
            "L": "CA",
            "ST": "San Francisco"
        }
    ]
}

重新生成member1证书：

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=peer member1.json | cfssljson -bare member1

验证证书：

openssl x509 -in member1.pem -text -noout

修改etcd1的etcd配置：

mv etcd.yaml ..
rm /var/lib/etcd/ -rf # 因为这是个扩展节点，需要同步etcd0的数据，所以把它自己数据删掉
vim ../etcd.yaml

修改后yaml文件u

apiVersion: v1
kind: Pod
metadata:
  annotations:
    scheduler.alpha.kubernetes.io/critical-pod: ""
  creationTimestamp: null
  labels:
    component: etcd
    tier: control-plane
  name: etcd
  namespace: kube-system
spec:
  containers:
  - command:
    - etcd
    - --advertise-client-urls=https://10.1.86.202:2379
    - --cert-file=/etc/kubernetes/pki/etcd/server.pem
    - --data-dir=/var/lib/etcd
    - --initial-advertise-peer-urls=https://10.1.86.202:2380
    - --initial-cluster=etcd0=https://10.1.86.201:2380,etcd1=https://10.1.86.202:2380
    - --key-file=/etc/kubernetes/pki/etcd/server-key.pem
    - --listen-client-urls=https://10.1.86.202:2379
    - --listen-peer-urls=https://10.1.86.202:2380
    - --name=etcd1
    - --peer-cert-file=/etc/kubernetes/pki/etcd/member1.pem
    - --peer-client-cert-auth=true
    - --peer-key-file=/etc/kubernetes/pki/etcd/member1-key.pem
    - --peer-trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    - --snapshot-count=10000
    - --trusted-ca-file=/etc/kubernetes/pki/etcd/ca.pem
    - --initial-cluster-state=existing  # 千万别加双引号，被坑死
    image: k8s.gcr.io/etcd-amd64:3.2.18
    imagePullPolicy: IfNotPresent
  # livenessProbe:
  #   exec:
  #     command:
  #     - /bin/sh
  #     - -ec
  #     - ETCDCTL_API=3 etcdctl --endpoints=https://[10.1.86.202]:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt
  #       --cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key
  #       get foo
  #   failureThreshold: 8
  #   initialDelaySeconds: 15
  #   timeoutSeconds: 15
    name: etcd
    resources: {}
    volumeMounts:
    - mountPath: /var/lib/etcd
      name: etcd-data
    - mountPath: /etc/kubernetes/pki/etcd
      name: etcd-certs
  hostNetwork: true
  priorityClassName: system-cluster-critical
  volumes:
  - hostPath:
      path: /var/lib/etcd
      type: DirectoryOrCreate
    name: etcd-data
  - hostPath:
      path: /etc/kubernetes/pki/cfssl
      type: DirectoryOrCreate
    name: etcd-certs
status: {}

在容器内查看集群已经健康运行了：

/ # alias etcdv2="ETCDCTL_API=2 etcdctl --endpoints=https://[10.1.86.201]:2379 --ca-file=/etc/kubernetes/pki/etcd/ca.pem --cert-file=/etc/kubernetes/pki/etcd/client.pem --key-file=/etc/kubernetes/pki/etcd/client-key.pem"
/ # etcdv2 cluster-health
member a874c87fd42044f is healthy: got healthy result from https://10.1.86.201:2379
member bbbbf223ec75e000 is healthy: got healthy result from https://10.1.86.202:2379
cluster is healthy

然后就可以把apiserver启动参数再加一个etcd1:

    - --etcd-servers=https://10.1.86.201:2379
    - --etcd-servers=https://10.1.86.202:2379

第三个节点同第二个，不再赘述。

细节问题非常多，一个端口，一个IP都不要填错，否则就会各种错误, 包括新加节点要清etcd数据这些小细节问题。
大功告成！