前言
环境准备
安装
静态集群
- 配置
- 启动测试
  - 查看集群状态
生成TLS证书
- etcd证书创建
静态TLS集群
- etcd 配置
- 启动测试
  - 检查TLS集群状态
ETCD 动态集群基于DNS的SRV解析自动发现
ETCD TLS动态集群基于DNS的SRV解析自动发现
报错解决
- 1. 证书报错 bad certificate
  - 解决
- 2. DNS 的 SRV 解析报错 cannot find local etcd member "etcd1" in SRV records
  - http不带证书解析如下
  - https带证书解析如下

前言

Etcd 是 CoreOS 基于 Raft 开发的分布式 key-value 存储，可用于服务发现、共享配置以及一致性保障（如数据库选主、分布式锁等）

本次环境，是用于k8s集群，由于在二进制部署 k8s 中，由于 Etcd 集群导致各种各样的问题，特意抽出时间来研究 Etcd 集群。

Etcd 集群配置分为三种:

静态发现
Etcd 动态发现
DNS 动态发现通过DNS的SRV解析动态发现集群

本次主要基于静态发现和 DNS动态发现两种，并结合自签的TLS证书来创建集群。

环境准备

此环境实际用于 k8s 中的ETCD集群使用，用于本次文档

主机名	角色	IP	系统版本	内核版本
node01.k8s.com	node01	192.168.1.91	CentOS 7.7	5.1.4-1.el7.elrepo.x86_64
node02.k8s.com	node02	192.168.1.92	CentOS 7.7	5.1.4-1.el7.elrepo.x86_64
node03.k8s.com	node03	192.168.1.93	CentOS 7.7	5.1.4-1.el7.elrepo.x86_64

安装

在三台机器上均执行

[root@node01 ~]# yum install etcd -y
[root@node01 ~]# rpm -qa etcd
etcd-3.3.11-2.el7.centos.x86_64

创建Etcd所需目录，在三台机器上均执行

mkdir /data/k8s/etcd/{data,wal} -p
mkdir -p /etc/kubernetes/cert
chown -R etcd.etcd /data/k8s/etcd

静态集群

配置

node01 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="http://192.168.1.91:2380"
ETCD_LISTEN_CLIENT_URLS="http://192.168.1.91:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd1"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="http://192.168.1.91:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://192.168.1.91:2379"

ETCD_INITIAL_CLUSTER="etcd1=http://192.168.1.91:2380,etcd2=http://192.168.1.92:2380,etcd3=http://192.168.1.93:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

node02 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="http://192.168.1.92:2380"
ETCD_LISTEN_CLIENT_URLS="http://192.168.1.92:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd2"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="http://192.168.1.92:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://192.168.1.92:2379"

ETCD_INITIAL_CLUSTER="etcd1=http://192.168.1.91:2380,etcd2=http://192.168.1.92:2380,etcd3=http://192.168.1.93:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

node03 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="http://192.168.1.93:2380"
ETCD_LISTEN_CLIENT_URLS="http://192.168.1.93:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd3"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="http://192.168.1.93:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://192.168.1.93:2379"

ETCD_INITIAL_CLUSTER="etcd1=http://192.168.1.91:2380,etcd2=http://192.168.1.92:2380,etcd3=http://192.168.1.93:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

启动测试

[root@node01 etcd]# systemctl start etcd
[root@node01 etcd]# systemctl status etcd
● etcd.service - Etcd Server
   Loaded: loaded (/usr/lib/systemd/system/etcd.service; disabled; vendor preset: disabled)
   Active: active (running) since Thu 2019-11-07 09:28:54 CST; 5s ago
 Main PID: 1546 (etcd)
    Tasks: 8
   Memory: 41.3M
   CGroup: /system.slice/etcd.service
           └─1546 /usr/bin/etcd --name=etcd1 --data-dir=/data/k8s/etcd/data --listen-client-urls=http://192.168.1.91:2379

Nov 07 09:28:54 node01.k8s.com etcd[1546]: 3b8b38de05e2c497 [term: 1] received a MsgVote message with higher term from 9c64fba479c5e94 [term: 2]
Nov 07 09:28:54 node01.k8s.com etcd[1546]: 3b8b38de05e2c497 became follower at term 2
Nov 07 09:28:54 node01.k8s.com etcd[1546]: 3b8b38de05e2c497 [logterm: 1, index: 3, vote: 0] cast MsgVote for 9c64fba479c5e94 [logterm: 1, index: 3] at term 2
Nov 07 09:28:54 node01.k8s.com etcd[1546]: raft.node: 3b8b38de05e2c497 elected leader 9c64fba479c5e94 at term 2
Nov 07 09:28:54 node01.k8s.com etcd[1546]: published {Name:etcd1 ClientURLs:[http://192.168.1.91:2379]} to cluster 19456f0bfd57284e
Nov 07 09:28:54 node01.k8s.com etcd[1546]: ready to serve client requests
Nov 07 09:28:54 node01.k8s.com etcd[1546]: serving insecure client requests on 192.168.1.91:2379, this is strongly discouraged!
Nov 07 09:28:54 node01.k8s.com systemd[1]: Started Etcd Server.
Nov 07 09:28:54 node01.k8s.com etcd[1546]: set the initial cluster version to 3.3
Nov 07 09:28:54 node01.k8s.com etcd[1546]: enabled capabilities for version 3.3

查看 /var/log/message 日志中，会有日下体现：

Nov  7 09:28:53 node02 etcd: added member 9c64fba479c5e94 [http://192.168.1.92:2380] to cluster 19456f0bfd57284e
Nov  7 09:28:53 node02 etcd: added member 3b8b38de05e2c497 [http://192.168.1.91:2380] to cluster 19456f0bfd57284e
Nov  7 09:28:53 node02 etcd: added member 76ea8679db7365b3 [http://192.168.1.93:2380] to cluster 19456f0bfd57284e

查看集群状态

[root@node01 etcd]# ETCDCTL_API=3 etcdctl --endpoints=http://192.168.1.91:2379,http://192.168.1.92:2379,http://192.168.1.93:2379 endpoint health
http://192.168.1.92:2379 is healthy: successfully committed proposal: took = 1.103545ms
http://192.168.1.93:2379 is healthy: successfully committed proposal: took = 2.122478ms
http://192.168.1.91:2379 is healthy: successfully committed proposal: took = 2.690215ms

[root@node01 etcd]# etcdctl --endpoints=http://192.168.1.91:2379,http://192.168.1.92:2379,http://192.168.1.93:2379 cluster-health
member 9c64fba479c5e94 is healthy: got healthy result from http://192.168.1.92:2379
member 3b8b38de05e2c497 is healthy: got healthy result from http://192.168.1.91:2379
member 76ea8679db7365b3 is healthy: got healthy result from http://192.168.1.93:2379
cluster is healthy

生成TLS证书

使用自签证书

CA(Certificate Authority)是自签名的根证书，用来签名后续创建的其他证书。本文章使用CloudFlare的PKI工具cfssl创建所有证书。

etcd证书创建

整个证书的创建过程均在 node01 上操作;

安装cfssl工具集

mkdir -p /opt/k8s/cert && cd /opt/k8s
wget https://pkg.cfssl.org/R1.2/cfssl_linux-amd64
mv cfssl_linux-amd64 /opt/k8s/bin/cfssl
wget https://pkg.cfssl.org/R1.2/cfssljson_linux-amd64
mv cfssljson_linux-amd64 /opt/k8s/bin/cfssljson
wget https://pkg.cfssl.org/R1.2/cfssl-certinfo_linux-amd64
mv cfssl-certinfo_linux-amd64 /opt/k8s/bin/cfssl-certinfo
chmod +x /opt/k8s/bin/*
echo 'export PATH=/opt/k8s/bin:$PATH' >> ~/.bash_profile
source ~/.bash_profile

生成证书

创建根证书 (CA)

CA证书是集群所有节点共享的，只需要创建一个CA证书，后续创建的所有证书都是由它签名

创建配置文件

CA配置文件用于配置根证书的使用场景(profile)和具体参数

(usage、过期时间、服务端认证、客户端认证、加密等)

cd /opt/k8s/work
cat > ca-config.json <<EOF
{
  "signing": {
    "default": {
      "expiry": "87600h"
    },
    "profiles": {
      "kubernetes": {
        "usages": [
            "signing",
            "key encipherment",
            "server auth",
            "client auth"
        ],
        "expiry": "87600h"
      }
    }
  }
}
EOF


######################
signing 表示该证书可用于签名其它证书，生成的ca.pem证书找中CA=TRUE
server auth 表示client可以用该证书对server提供的证书进行验证
client auth 表示server可以用该证书对client提供的证书进行验证

创建证书签名请求文件

cd /opt/k8s/work
cat > ca-csr.json <<EOF
{
  "CN": "kubernetes",
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "ST": "BeiJing",
      "L": "BeiJing",
      "O": "k8s",
      "OU": "4Paradigm"
    }
  ],
  "ca": {
    "expiry": "876000h"
 }
}
EOF



#######################
CN CommonName,kube-apiserver从证书中提取该字段作为请求的用户名(User Name)，浏览器使用该字段验证网站是否合法
O Organization,kube-apiserver 从证书中提取该字段作为请求用户和所属组(Group)
kube-apiserver将提取的User、Group作为RBAC授权的用户和标识

生成CA证书和私钥

cd /opt/k8s/work
cfssl gencert -initca ca-csr.json | cfssljson -bare ca
ls ca*

创建etcd证书和私钥

cd /opt/k8s/work
cat > etcd-csr.json <<EOF
{
  "CN": "etcd",
  "hosts": [
    "127.0.0.1",
    "192.168.1.91",
    "192.168.1.92",
    "192.168.1.93",
    "k8s.com",
    "etcd1.k8s.com",
    "etcd2.k8s.com",
    "etcd3.k8s.com"
  ],
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "ST": "BeiJing",
      "L": "BeiJing",
      "O": "k8s",
      "OU": "4Paradigm"
    }
  ]
}
EOF

#host字段指定授权使用该证书的etcd节点IP或域名列表，需要将etcd集群的3个节点都添加其中

生成证书和私钥

cd /opt/k8s/work

cfssl gencert -ca=/opt/k8s/work/ca.pem \
    -ca-key=/opt/k8s/work/ca-key.pem \
    -config=/opt/k8s/work/ca-config.json \
    -profile=kubernetes etcd-csr.json | cfssljson -bare etcd
    
ls etcd*pem -l
-rw------- 1 root root 1675 Nov  7 09:52 etcd-key.pem
-rw-r--r-- 1 root root 1444 Nov  7 09:52 etcd.pem

etcd 使用的TLS证书创建完成

分发证书到各节点上

要做所有节点上创建对应的目录

mkdir /data/k8s/etcd/{data,wal} -p
mkdir -p /etc/kubernetes/cert
chown -R etcd.etcd /data/k8s/etcd

分发证书

cd /opt/k8s/work
scp ca*.pem ca-config.json 192.168.1.91:/etc/kubernetes/cert
scp ca*.pem ca-config.json 192.168.1.92:/etc/kubernetes/cert 
scp ca*.pem ca-config.json 192.168.1.93:/etc/kubernetes/cert
scp etcd*pem 192.168.1.91:/etc/etcd/cert/
scp etcd*pem 192.168.1.92:/etc/etcd/cert/
scp etcd*pem 192.168.1.93:/etc/etcd/cert/

在所有节点上执行:

chown -R etcd.etcd /etc/etcd/cert

静态TLS集群

etcd 配置

node01 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="https://192.168.1.91:2380"
ETCD_LISTEN_CLIENT_URLS="https://192.168.1.91:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd1"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="https://192.168.1.91:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://192.168.1.91:2379"
ETCD_INITIAL_CLUSTER="etcd1=https://192.168.1.91:2380,etcd2=https://192.168.1.92:2380,etcd3=https://192.168.1.93:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

ETCD_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_CLIENT_CERT_AUTH="true"
ETCD_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_AUTO_TLS="true"
ETCD_PEER_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_PEER_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_PEER_CLIENT_CERT_AUTH="true"
ETCD_PEER_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_PEER_AUTO_TLS="true"

node02 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="https://192.168.1.92:2380"
ETCD_LISTEN_CLIENT_URLS="https://192.168.1.92:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd2"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="https://192.168.1.92:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://192.168.1.92:2379"
ETCD_INITIAL_CLUSTER="etcd1=https://192.168.1.91:2380,etcd2=https://192.168.1.92:2380,etcd3=https://192.168.1.93:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

ETCD_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_CLIENT_CERT_AUTH="true"
ETCD_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_AUTO_TLS="true"
ETCD_PEER_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_PEER_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_PEER_CLIENT_CERT_AUTH="true"
ETCD_PEER_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_PEER_AUTO_TLS="true"

node03 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="https://192.168.1.93:2380"
ETCD_LISTEN_CLIENT_URLS="https://192.168.1.93:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd3"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="https://192.168.1.93:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://192.168.1.93:2379"
ETCD_INITIAL_CLUSTER="etcd1=https://192.168.1.91:2380,etcd2=https://192.168.1.92:2380,etcd3=https://192.168.1.93:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

ETCD_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_CLIENT_CERT_AUTH="true"
ETCD_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_AUTO_TLS="true"
ETCD_PEER_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_PEER_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_PEER_CLIENT_CERT_AUTH="true"
ETCD_PEER_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_PEER_AUTO_TLS="true"

启动测试

[root@node01 work]# systemctl start etcd
[root@node01 work]# systemctl status etcd
● etcd.service - Etcd Server
   Loaded: loaded (/usr/lib/systemd/system/etcd.service; disabled; vendor preset: disabled)
   Active: active (running) since Thu 2019-11-07 10:15:58 CST; 5s ago
 Main PID: 2078 (etcd)
    Tasks: 8
   Memory: 28.9M
   CGroup: /system.slice/etcd.service
           └─2078 /usr/bin/etcd --name=etcd1 --data-dir=/data/k8s/etcd/data --listen-client-urls=https://192.168.1.91:2379

Nov 07 10:15:58 node01.k8s.com etcd[2078]: 2a40d8ba966d12fe [term: 1] received a MsgVote message with higher term from af05139f75a68867 [term: 2]
Nov 07 10:15:58 node01.k8s.com etcd[2078]: 2a40d8ba966d12fe became follower at term 2
Nov 07 10:15:58 node01.k8s.com etcd[2078]: 2a40d8ba966d12fe [logterm: 1, index: 3, vote: 0] cast MsgVote for af05139f75a68867 [logterm: 1, index: 3] at term 2
Nov 07 10:15:58 node01.k8s.com etcd[2078]: raft.node: 2a40d8ba966d12fe elected leader af05139f75a68867 at term 2
Nov 07 10:15:58 node01.k8s.com etcd[2078]: published {Name:etcd1 ClientURLs:[https://192.168.1.91:2379]} to cluster f3e9c54e1aafb3c1
Nov 07 10:15:58 node01.k8s.com etcd[2078]: ready to serve client requests
Nov 07 10:15:58 node01.k8s.com etcd[2078]: serving client requests on 192.168.1.91:2379
Nov 07 10:15:58 node01.k8s.com systemd[1]: Started Etcd Server.
Nov 07 10:15:58 node01.k8s.com etcd[2078]: set the initial cluster version to 3.3
Nov 07 10:15:58 node01.k8s.com etcd[2078]: enabled capabilities for version 3.3

查看 /var/log/message 日志中，会有日下体现：

Nov  7 10:15:57 node01 etcd: added member 2a40d8ba966d12fe [https://192.168.1.91:2380] to cluster f3e9c54e1aafb3c1
Nov  7 10:15:57 node01 etcd: added member af05139f75a68867 [https://192.168.1.92:2380] to cluster f3e9c54e1aafb3c1
Nov  7 10:15:57 node01 etcd: added member c3bab7c20fba3f60 [https://192.168.1.93:2380] to cluster f3e9c54e1aafb3c1

检查TLS集群状态

ETCDCTL_API=3 etcdctl \
    --endpoints=https://etcd1.k8s.com:2379,https://etcd2.k8s.com:2379,https://etcd3.k8s.com:2379 \
    --cacert=/etc/kubernetes/cert/ca.pem \
    --cert=/etc/etcd/cert/etcd.pem \
    --key=/etc/etcd/cert/etcd-key.pem endpoint health

# 输出
https://192.168.1.92:2379 is healthy: successfully committed proposal: took = 1.317022ms
https://192.168.1.91:2379 is healthy: successfully committed proposal: took = 1.59958ms
https://192.168.1.93:2379 is healthy: successfully committed proposal: took = 1.453049ms

etcdctl \
--endpoint=https://etcd1.k8s.com:2379 \
--ca-file=/etc/kubernetes/cert/ca.pem \
--cert-file=/etc/etcd/cert/etcd.pem \
--key-file=/etc/etcd/cert/etcd-key.pem cluster-health

# 输出
member 40a8f19a5db99534 is healthy: got healthy result from https://etcd2.k8s.com:2379
member 9888555207dbf0e0 is healthy: got healthy result from https://etcd3.k8s.com:2379
member a0d541999e9eb3b3 is healthy: got healthy result from https://etcd1.k8s.com:2379
cluster is healthy

ETCD 动态集群基于DNS的SRV解析自动发现

需要局域网内部有DNS服务器

添加SRV解析

目前常用的内部DNS服务有两种，bind、dnsmasq

在下面都会列出具体的配置，但只需要配置其中之一即可；

方法一: 使用`bind`配置SRV解析

如果内部没有bind服务，可以参考部署文档文章: https://www.cnblogs.com/winstom/p/11806962.html

使用域名为 : k8s.com，在bind的zone文件中添加如下解析:

etcd1   IN      A       192.168.1.91
etcd2   IN      A       192.168.1.92
etcd3   IN      A       192.168.1.93
_etcd-server._tcp.k8s.com.  IN      SRV     10    10       2380     etcd1
_etcd-server._tcp.k8s.com.  IN      SRV     10    10       2380     etcd2
_etcd-server._tcp.k8s.com.  IN      SRV     10    10       2380     etcd3
_etcd-client._tcp.k8s.com.  IN      SRV     10    10       2379     etcd1
_etcd-client._tcp.k8s.com.  IN      SRV     10    10       2379     etcd2
_etcd-client._tcp.k8s.com.  IN      SRV     10    10       2379     etcd3

修改之后重新加载配置文件:

[root@jenkins named]# named-checkzone k8s.com k8s.com.zone 
zone k8s.com/IN: loaded serial 0
OK
[root@jenkins named]# rndc reload
server reload successful

方法二: 使用`dnsmasq`配置SRV解析

如果内部没有dnsmasq服务，可以参考部署文档文章: https://www.cnblogs.com/winstom/p/11809066.html

使用域名为 : k8s.com，具体修改如下:

在/etc/dnsmasq_hosts新增下面内容

192.168.1.91 etcd1 etcd1.k8s.com
192.168.1.92 etcd2 etcd2.k8s.com
192.168.1.93 etcd3 etcd3.k8s.com

在 /etc/dnsmasq.conf 文件中增加下面SRV解析内容

srv-host=_etcd-server._tcp.k8s.com,etcd1.k8s.com,2380,0,100
srv-host=_etcd-server._tcp.k8s.com,etcd2.k8s.com,2380,0,100
srv-host=_etcd-server._tcp.k8s.com,etcd3.k8s.com,2380,0,100
srv-host=_etcd-client._tcp.k8s.com,etcd1.k8s.com,2379,0,100
srv-host=_etcd-client._tcp.k8s.com,etcd2.k8s.com,2379,0,100
srv-host=_etcd-client._tcp.k8s.com,etcd3.k8s.com,2379,0,100

修改之后重启服务 systemctl restart dnsmasq

验证SRV解析是否正常

查询SRV记录

[root@node01 ~]# dig @192.168.1.122 +noall +answer SRV _etcd-server._tcp.k8s.com
_etcd-server._tcp.k8s.com. 3600 IN  SRV 2380 0 100 etcd2.k8s.com.
_etcd-server._tcp.k8s.com. 3600 IN  SRV 2380 0 100 etcd1.k8s.com.
_etcd-server._tcp.k8s.com. 3600 IN  SRV 2380 0 100 etcd3.k8s.com.

查询域名解析结果

[root@node01 ~]# dig @192.168.1.122 +noall +answer etcd1.k8s.com etcd2.k8s.com etcd3.k8s.com
etcd1.k8s.com.      86400   IN  A   192.168.1.91
etcd2.k8s.com.      86400   IN  A   192.168.1.92
etcd3.k8s.com.      86400   IN  A   192.168.1.93

如上述显示，则表示SRV解析正常

配置ETCD

node01 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="http://192.168.1.91:2380"
ETCD_LISTEN_CLIENT_URLS="http://192.168.1.91:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd1"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="http://etcd1.k8s.com:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://etcd1.k8s.com:2379"
ETCD_DISCOVERY_SRV="k8s.com"

ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

node02 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="http://192.168.1.92:2380"
ETCD_LISTEN_CLIENT_URLS="http://192.168.1.92:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd2"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="http://etcd2.k8s.com:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://etcd2.k8s.com:2379"
ETCD_DISCOVERY_SRV="k8s.com"

ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

node03 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="http://192.168.1.93:2380"
ETCD_LISTEN_CLIENT_URLS="http://192.168.1.93:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd3"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="http://etcd3.k8s.com:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://etcd3.k8s.com:2379"
ETCD_DISCOVERY_SRV="k8s.com"

ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

启动并测试

启动

[root@node01 etcd]# systemctl start etcd
[root@node01 etcd]# systemctl status etcd
● etcd.service - Etcd Server
   Loaded: loaded (/usr/lib/systemd/system/etcd.service; disabled; vendor preset: disabled)
   Active: active (running) since Thu 2019-11-07 11:25:29 CST; 4s ago
 Main PID: 14203 (etcd)
    Tasks: 8
   Memory: 16.9M
   CGroup: /system.slice/etcd.service
           └─14203 /usr/bin/etcd --name=etcd1 --data-dir=/data/k8s/etcd/data --listen-client-urls=http://192.168.1.91:2379

Nov 07 11:25:29 node01.k8s.com etcd[14203]: d79e9ae86b2a1de1 [quorum:2] has received 2 MsgVoteResp votes and 0 vote rejections
Nov 07 11:25:29 node01.k8s.com etcd[14203]: d79e9ae86b2a1de1 became leader at term 2
Nov 07 11:25:29 node01.k8s.com etcd[14203]: raft.node: d79e9ae86b2a1de1 elected leader d79e9ae86b2a1de1 at term 2
Nov 07 11:25:29 node01.k8s.com etcd[14203]: published {Name:etcd1 ClientURLs:[http://etcd1.k8s.com:2379 http://etcd1.k8s.com:4001]} to cluster 42cecf80e3791d6c
Nov 07 11:25:29 node01.k8s.com etcd[14203]: ready to serve client requests
Nov 07 11:25:29 node01.k8s.com etcd[14203]: serving insecure client requests on 192.168.1.91:2379, this is strongly discouraged!
Nov 07 11:25:29 node01.k8s.com systemd[1]: Started Etcd Server.
Nov 07 11:25:29 node01.k8s.com etcd[14203]: setting up the initial cluster version to 3.3
Nov 07 11:25:29 node01.k8s.com etcd[14203]: set the initial cluster version to 3.3
Nov 07 11:25:29 node01.k8s.com etcd[14203]: enabled capabilities for version 3.3

日志 vim /var/log/messages 表现如下:

Nov  7 11:25:27 node01 etcd: got bootstrap from DNS for etcd-server at 0=http://etcd3.k8s.com:2380
Nov  7 11:25:27 node01 etcd: got bootstrap from DNS for etcd-server at 1=http://etcd2.k8s.com:2380
Nov  7 11:25:27 node01 etcd: got bootstrap from DNS for etcd-server at etcd1=http://etcd1.k8s.com:2380
Nov  7 11:25:27 node01 etcd: resolving etcd1.k8s.com:2380 to 192.168.1.91:2380
Nov  7 11:25:27 node01 etcd: resolving etcd1.k8s.com:2380 to 192.168.1.91:2380
Nov  7 11:25:28 node01 etcd: name = etcd1
Nov  7 11:25:28 node01 etcd: data dir = /data/k8s/etcd/data
Nov  7 11:25:28 node01 etcd: member dir = /data/k8s/etcd/data/member
Nov  7 11:25:28 node01 etcd: dedicated WAL dir = /data/k8s/etcd/wal
Nov  7 11:25:28 node01 etcd: heartbeat = 100ms
Nov  7 11:25:28 node01 etcd: election = 1000ms
Nov  7 11:25:28 node01 etcd: snapshot count = 100000
Nov  7 11:25:28 node01 etcd: advertise client URLs = http://etcd1.k8s.com:2379,http://etcd1.k8s.com:4001
Nov  7 11:25:28 node01 etcd: initial advertise peer URLs = http://etcd1.k8s.com:2380
Nov  7 11:25:28 node01 etcd: initial cluster = 0=http://etcd3.k8s.com:2380,1=http://etcd2.k8s.com:2380,etcd1=http://etcd1.k8s.com:2380

测试：

[root@node01 etcd]# etcdctl --endpoints=http://192.168.1.91:2379 cluster-health
member 184beca37ca32d75 is healthy: got healthy result from http://etcd2.k8s.com:2379
member d79e9ae86b2a1de1 is healthy: got healthy result from http://etcd1.k8s.com:2379
member f7662e609b7e4013 is healthy: got healthy result from http://etcd3.k8s.com:2379
cluster is healthy

ETCD TLS动态集群基于DNS的SRV解析自动发现

需要局域网内部有DNS服务器

添加SRV解析

目前常用的内部DNS服务有两种，bind、dnsmasq

在下面都会列出具体的配置，但只需要配置其中之一即可；

方法一: 使用`bind`配置SRV解析

如果内部没有bind服务，可以参考部署文档文章: https://www.cnblogs.com/winstom/p/11806962.html

使用域名为 : k8s.com，在bind的zone文件中添加如下解析:

etcd1   IN      A       192.168.1.91
etcd2   IN      A       192.168.1.92
etcd3   IN      A       192.168.1.93
_etcd-server-ssl._tcp.k8s.com.  IN      SRV     10    10       2380     etcd1
_etcd-server-ssl._tcp.k8s.com.  IN      SRV     10    10       2380     etcd2
_etcd-server-ssl._tcp.k8s.com.  IN      SRV     10    10       2380     etcd3
_etcd-client-ssl._tcp.k8s.com.  IN      SRV     10    10       2379     etcd1
_etcd-client-ssl._tcp.k8s.com.  IN      SRV     10    10       2379     etcd2
_etcd-client-ssl._tcp.k8s.com.  IN      SRV     10    10       2379     etcd3

修改之后重新加载配置文件:

[root@jenkins named]# named-checkzone k8s.com k8s.com.zone 
zone k8s.com/IN: loaded serial 0
OK
[root@jenkins named]# rndc reload
server reload successful

方法二: 使用`dnsmasq`配置SRV解析

如果内部没有dnsmasq服务，可以参考部署文档文章: https://www.cnblogs.com/winstom/p/11809066.html

使用域名为 : k8s.com，具体修改如下:

在/etc/dnsmasq_hosts新增下面内容

192.168.1.91 etcd1 etcd1.k8s.com
192.168.1.92 etcd2 etcd2.k8s.com
192.168.1.93 etcd3 etcd3.k8s.com

在 /etc/dnsmasq.conf 文件中增加下面SRV解析内容

srv-host=_etcd-server-ssl._tcp.k8s.com,etcd1.k8s.com,2380,0,100
srv-host=_etcd-server-ssl._tcp.k8s.com,etcd2.k8s.com,2380,0,100
srv-host=_etcd-server-ssl._tcp.k8s.com,etcd3.k8s.com,2380,0,100
srv-host=_etcd-client-ssl._tcp.k8s.com,etcd1.k8s.com,2379,0,100
srv-host=_etcd-client-ssl._tcp.k8s.com,etcd2.k8s.com,2379,0,100
srv-host=_etcd-client-ssl._tcp.k8s.com,etcd3.k8s.com,2379,0,100

修改之后重启服务 systemctl restart dnsmasq

验证SRV解析是否正常

查询SRV记录

[root@node01 etcd]# dig @192.168.1.122 +noall +answer SRV _etcd-server-ssl._tcp.k8s.com
_etcd-server-ssl._tcp.k8s.com. 3600 IN  SRV 2380 0 100 etcd3.k8s.com.
_etcd-server-ssl._tcp.k8s.com. 3600 IN  SRV 2380 0 100 etcd2.k8s.com.
_etcd-server-ssl._tcp.k8s.com. 3600 IN  SRV 2380 0 100 etcd1.k8s.com.

查询域名解析结果

[root@node01 ~]# dig @192.168.1.122 +noall +answer etcd1.k8s.com etcd2.k8s.com etcd3.k8s.com
etcd1.k8s.com.      86400   IN  A   192.168.1.91
etcd2.k8s.com.      86400   IN  A   192.168.1.92
etcd3.k8s.com.      86400   IN  A   192.168.1.93

ETCD 配置

node01 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="https://192.168.1.91:2380"
ETCD_LISTEN_CLIENT_URLS="https://192.168.1.91:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd1"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="https://etcd1.k8s.com:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://etcd1.k8s.com:2379,https://etcd1.k8s.com:4001"
ETCD_DISCOVERY_SRV="k8s.com"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

ETCD_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_CLIENT_CERT_AUTH="true"
ETCD_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_AUTO_TLS="true"
ETCD_PEER_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_PEER_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_PEER_CLIENT_CERT_AUTH="true"
ETCD_PEER_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_PEER_AUTO_TLS="true"

node02 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="https://192.168.1.92:2380"
ETCD_LISTEN_CLIENT_URLS="https://192.168.1.92:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd2"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="https://etcd2.k8s.com:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://etcd2.k8s.com:2379"
ETCD_DISCOVERY_SRV="k8s.com"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

ETCD_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_CLIENT_CERT_AUTH="true"
ETCD_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_AUTO_TLS="true"
ETCD_PEER_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_PEER_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_PEER_CLIENT_CERT_AUTH="true"
ETCD_PEER_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_PEER_AUTO_TLS="true"

node03 配置文件

ETCD_DATA_DIR="/data/k8s/etcd/data"
ETCD_WAL_DIR="/data/k8s/etcd/wal"
ETCD_LISTEN_PEER_URLS="https://192.168.1.93:2380"
ETCD_LISTEN_CLIENT_URLS="https://192.168.1.93:2379"
ETCD_MAX_SNAPSHOTS="5"
ETCD_MAX_WALS="5"
ETCD_NAME="etcd3"
ETCD_SNAPSHOT_COUNT="100000"
ETCD_HEARTBEAT_INTERVAL="100"
ETCD_ELECTION_TIMEOUT="1000"

ETCD_INITIAL_ADVERTISE_PEER_URLS="https://etcd3.k8s.com:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://etcd3.k8s.com:2379"
ETCD_DISCOVERY_SRV="k8s.com"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

ETCD_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_CLIENT_CERT_AUTH="true"
ETCD_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_AUTO_TLS="true"
ETCD_PEER_CERT_FILE="/etc/etcd/cert/etcd.pem"
ETCD_PEER_KEY_FILE="/etc/etcd/cert/etcd-key.pem"
ETCD_PEER_CLIENT_CERT_AUTH="true"
ETCD_PEER_TRUSTED_CA_FILE="/etc/kubernetes/cert/ca.pem"
ETCD_PEER_AUTO_TLS="true"

启动测试

启动

[root@node03 etcd]# systemctl restart etcd
[root@node03 etcd]# systemctl status etcd
● etcd.service - Etcd Server
   Loaded: loaded (/usr/lib/systemd/system/etcd.service; disabled; vendor preset: disabled)
   Active: active (running) since Thu 2019-11-07 12:38:37 CST; 4s ago
 Main PID: 13460 (etcd)
    Tasks: 8
   Memory: 16.6M
   CGroup: /system.slice/etcd.service
           └─13460 /usr/bin/etcd --name=etcd3 --data-dir=/data/k8s/etcd/data --listen-client-urls=https://192.168.1.93:2379

Nov 07 12:38:36 node03.k8s.com etcd[13460]: established a TCP streaming connection with peer 40a8f19a5db99534 (stream Message writer)
Nov 07 12:38:36 node03.k8s.com etcd[13460]: established a TCP streaming connection with peer 40a8f19a5db99534 (stream MsgApp v2 writer)
Nov 07 12:38:37 node03.k8s.com etcd[13460]: 9888555207dbf0e0 [term: 92] received a MsgVote message with higher term from a0d541999e9eb3b3 [term: 98]
Nov 07 12:38:37 node03.k8s.com etcd[13460]: 9888555207dbf0e0 became follower at term 98
Nov 07 12:38:37 node03.k8s.com etcd[13460]: 9888555207dbf0e0 [logterm: 92, index: 9, vote: 0] cast MsgVote for a0d541999e9eb3b3 [logterm: 92, index: 9] at term 98
Nov 07 12:38:37 node03.k8s.com etcd[13460]: raft.node: 9888555207dbf0e0 elected leader a0d541999e9eb3b3 at term 98
Nov 07 12:38:37 node03.k8s.com etcd[13460]: published {Name:etcd3 ClientURLs:[https://etcd3.k8s.com:2379]} to cluster f445a02ce3dc6a02
Nov 07 12:38:37 node03.k8s.com etcd[13460]: ready to serve client requests
Nov 07 12:38:37 node03.k8s.com etcd[13460]: serving client requests on 192.168.1.93:2379
Nov 07 12:38:37 node03.k8s.com systemd[1]: Started Etcd Server.

日志体现

Nov  7 12:38:36 node01 etcd: added member 40a8f19a5db99534 [https://etcd2.k8s.com:2380] to cluster f445a02ce3dc6a02
Nov  7 12:38:36 node01 etcd: starting peer 40a8f19a5db99534...
Nov  7 12:38:36 node01 etcd: started HTTP pipelining with peer 40a8f19a5db99534
Nov  7 12:38:36 node01 etcd: started streaming with peer 40a8f19a5db99534 (writer)
Nov  7 12:38:36 node01 etcd: started peer 40a8f19a5db99534
Nov  7 12:38:36 node01 etcd: added peer 40a8f19a5db99534
Nov  7 12:38:36 node01 etcd: added member 9888555207dbf0e0 [https://etcd3.k8s.com:2380] to cluster f445a02ce3dc6a02
Nov  7 12:38:36 node01 etcd: starting peer 9888555207dbf0e0...
Nov  7 12:38:36 node01 etcd: started HTTP pipelining with peer 9888555207dbf0e0
Nov  7 12:38:36 node01 etcd: started peer 9888555207dbf0e0
Nov  7 12:38:36 node01 etcd: added peer 9888555207dbf0e0
Nov  7 12:38:36 node01 etcd: added member a0d541999e9eb3b3 [https://etcd1.k8s.com:2380] to cluster f445a02ce3dc6a02

测试集群状态：

ETCDCTL_API=3 etcdctl --endpoints=https://etcd1.k8s.com:2379,https://etcd2.k8s.com:2379,https://etcd3.k8s.com:2379 \
    --cacert=/etc/kubernetes/cert/ca.pem \
    --cert=/etc/etcd/cert/etcd.pem \
    --key=/etc/etcd/cert/etcd-key.pem endpoint health

# 输出
https://etcd1.k8s.com:2379 is healthy: successfully committed proposal: took = 4.269468ms
https://etcd3.k8s.com:2379 is healthy: successfully committed proposal: took = 1.58797ms
https://etcd2.k8s.com:2379 is healthy: successfully committed proposal: took = 1.622151ms

etcdctl \
--endpoint=https://etcd1.k8s.com:2379 \
--ca-file=/etc/kubernetes/cert/ca.pem \
--cert-file=/etc/etcd/cert/etcd.pem \
--key-file=/etc/etcd/cert/etcd-key.pem cluster-health

# 输出
member 40a8f19a5db99534 is healthy: got healthy result from https://etcd2.k8s.com:2379
member 9888555207dbf0e0 is healthy: got healthy result from https://etcd3.k8s.com:2379
member a0d541999e9eb3b3 is healthy: got healthy result from https://etcd1.k8s.com:2379
cluster is healthy

报错解决

1. 证书报错 bad certificate

日志中报错:

Nov  7 12:37:03 node01 etcd: rejected connection from "192.168.1.93:46294" (error "remote error: tls: bad certificate", ServerName "k8s.com")

解决

报错的意思是在生成ETCD的TLS证书的时候，没有把对应的域名加进去

在创建ETCD的TLS证书请求的文件中加入对应的域名

cd /opt/k8s/work
cat > etcd-csr.json <<EOF
{
  "CN": "etcd",
  "hosts": [
    "127.0.0.1",
    "192.168.1.91",
    "192.168.1.92",
    "192.168.1.93",
    "k8s.com",          # 这里的域名查看是否正确
    "etcd1.k8s.com",
    "etcd2.k8s.com",
    "etcd3.k8s.com"
  ],
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "ST": "BeiJing",
      "L": "BeiJing",
      "O": "k8s",
      "OU": "4Paradigm"
    }
  ]
}
EOF

2. `DNS` 的 `SRV` 解析报错 cannot find local etcd member "etcd1" in SRV records

DNS 如果配置有问题，会有如下报错:

etcd: error setting up initial cluster: cannot find local etcd member "etcd1" in SRV records

这里是表示DNS在配置SRV解析的时候报错，请仔细查看解析配置:

SRV解析分为两种，一种是http不带证书的解析，一种是https带证书的解析，是有区别的，如果配置错误就会包上述错误

http不带证书解析如下

bind 的解析

编辑 /var/named/k8s.com.zone 文件

etcd1   IN      A       192.168.1.91
etcd2   IN      A       192.168.1.92
etcd3   IN      A       192.168.1.93
_etcd-server._tcp.k8s.com.  IN      SRV     10    10       2380     etcd1
_etcd-server._tcp.k8s.com.  IN      SRV     10    10       2380     etcd2
_etcd-server._tcp.k8s.com.  IN      SRV     10    10       2380     etcd3
_etcd-client._tcp.k8s.com.  IN      SRV     10    10       2379     etcd1
_etcd-client._tcp.k8s.com.  IN      SRV     10    10       2379     etcd2
_etcd-client._tcp.k8s.com.  IN      SRV     10    10       2379     etcd3

dnsmasq 的解析

在/etc/dnsmasq_hosts新增下面内容

192.168.1.91 etcd1 etcd1.k8s.com
192.168.1.92 etcd2 etcd2.k8s.com
192.168.1.93 etcd3 etcd3.k8s.com

在 /etc/dnsmasq.conf 文件中增加下面SRV解析内容

srv-host=_etcd-server._tcp.k8s.com,etcd1.k8s.com,2380,0,100
srv-host=_etcd-server._tcp.k8s.com,etcd2.k8s.com,2380,0,100
srv-host=_etcd-server._tcp.k8s.com,etcd3.k8s.com,2380,0,100
srv-host=_etcd-client._tcp.k8s.com,etcd1.k8s.com,2380,0,100
srv-host=_etcd-client._tcp.k8s.com,etcd2.k8s.com,2380,0,100
srv-host=_etcd-client._tcp.k8s.com,etcd3.k8s.com,2380,0,100

https带证书解析如下

bind 的解析

编辑 /var/named/k8s.com.zone 文件

etcd1   IN      A       192.168.1.91
etcd2   IN      A       192.168.1.92
etcd3   IN      A       192.168.1.93
_etcd-server-ssl._tcp.k8s.com.  IN      SRV     10    10       2380     etcd1
_etcd-server-ssl._tcp.k8s.com.  IN      SRV     10    10       2380     etcd2
_etcd-server-ssl._tcp.k8s.com.  IN      SRV     10    10       2380     etcd3
_etcd-client-ssl._tcp.k8s.com.  IN      SRV     10    10       2379     etcd1
_etcd-client-ssl._tcp.k8s.com.  IN      SRV     10    10       2379     etcd2
_etcd-client-ssl._tcp.k8s.com.  IN      SRV     10    10       2379     etcd3

dnsmasq 的解析

在/etc/dnsmasq_hosts新增下面内容

192.168.1.91 etcd1 etcd1.k8s.com
192.168.1.92 etcd2 etcd2.k8s.com
192.168.1.93 etcd3 etcd3.k8s.com

在 /etc/dnsmasq.conf 文件中增加下面SRV解析内容

srv-host=_etcd-server-ssl._tcp.k8s.com,etcd1.k8s.com,2380,0,100
srv-host=_etcd-server-ssl._tcp.k8s.com,etcd2.k8s.com,2380,0,100
srv-host=_etcd-server-ssl._tcp.k8s.com,etcd3.k8s.com,2380,0,100
srv-host=_etcd-client-ssl._tcp.k8s.com,etcd1.k8s.com,2379,0,100
srv-host=_etcd-client-ssl._tcp.k8s.com,etcd2.k8s.com,2379,0,100
srv-host=_etcd-client-ssl._tcp.k8s.com,etcd3.k8s.com,2379,0,100

来源：https://www.cnblogs.com/winstom/p/11811373.html

标签

etcd

Kubernetes

CentOS

集群服务器

k8s

srv记录

CentOS 7 ETCD集群配置大全

前言

环境准备

安装

静态集群

配置

node01 配置文件

node02 配置文件

node03 配置文件

启动测试

查看集群状态

生成TLS证书

etcd证书创建

安装cfssl工具集

生成证书

分发证书到各节点上

静态TLS集群

etcd 配置

node01 配置文件

node02 配置文件

node03 配置文件

启动测试

检查TLS集群状态

ETCD 动态集群基于DNS的SRV解析自动发现

添加SRV解析

方法一: 使用bind配置SRV解析

方法二: 使用dnsmasq配置SRV解析

验证SRV解析是否正常

配置ETCD

node01 配置文件

node02 配置文件

node03 配置文件

启动并测试

ETCD TLS动态集群基于DNS的SRV解析自动发现

添加SRV解析

方法一: 使用bind配置SRV解析

方法二: 使用dnsmasq配置SRV解析

验证SRV解析是否正常

ETCD 配置

node01 配置文件

node02 配置文件

node03 配置文件

启动测试

报错解决

1. 证书报错 bad certificate

解决

2. DNS 的 SRV 解析报错 cannot find local etcd member "etcd1" in SRV records

http不带证书解析如下

bind 的解析

dnsmasq 的解析

https带证书解析如下

bind 的解析

dnsmasq 的解析

方法一: 使用`bind`配置SRV解析

方法二: 使用`dnsmasq`配置SRV解析

方法一: 使用`bind`配置SRV解析

方法二: 使用`dnsmasq`配置SRV解析

2. `DNS` 的 `SRV` 解析报错 cannot find local etcd member "etcd1" in SRV records