文章目录
- 1, 生成 TLS 秘钥对
- 2,拷贝密钥对到所有节点
- 3,配置 etcd 使用证书
- 4,测试 etcd 是否正常
- 5,配置 kube-apiserver 使用 CA 连接 etcd
- 6,测试 kube-apiserver
- 7,未解决的问题
SSL/TSL 认证分单向认证和双向认证两种方式。简单说就是单向认证只是客户端对服务端的身份进行验证,双向认证是客户端和服务端互相进行身份认证。就比如,我们登录淘宝买东西,为了防止我们登录的是假淘宝网站,此时我们通过浏览器打开淘宝买东西时,浏览器会验证我们登录的网站是否是真的淘宝的网站,而淘宝网站不关心我们是否“合法”,这就是单向认证。而双向认证是服务端也需要对客户端做出认证。
因为大部分 kubernetes 基于内网部署,而内网应该都会采用私有 IP 地址通讯,权威 CA 好像只能签署域名证书,对于签署到 IP 可能无法实现。所以我们需要预先自建 CA 签发证书。
Generate self-signed certificates 官方参考文档
官方推荐使用 cfssl 来自建 CA 签发证书,当然你也可以用众人熟知的 OpenSSL 或者 easy-rsa。以下步骤遵循官方文档:
1, 生成 TLS 秘钥对
生成步骤:
- 1,下载 cfssl
- 2,初始化证书颁发机构
- 3,配置 CA 选项
- 4,生成服务器端证书
- 5,生成对等证书
- 6,生成客户端证书
想深入了解 HTTPS 的看这里:
1,下载 cfssl
- 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
2,初始化证书颁发机构
- mkdir ~/cfssl
- cd ~/cfssl
- cfssl print-defaults config > ca-config.json
- cfssl print-defaults csr > ca-csr.json
证书类型介绍:
- client certificate 用于通过服务器验证客户端。例如etcdctl,etcd proxy,fleetctl或docker客户端。
- server certificate 由服务器使用,并由客户端验证服务器身份。例如docker服务器或kube-apiserver。
- peer certificate 由 etcd 集群成员使用,供它们彼此之间通信使用。
3,配置 CA 选项
- $ cat << EOF > 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"
- ]
- }
- }
- }
- }
- $ cat << EOF > 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"
- }
- ]
- }
- 生成 CA 证书:
- $ cfssl gencert -initca ca-csr.json | cfssljson -bare ca -
- 将会生成以下几个文件:
- ca-key.pem
- ca.csr
- ca.pem
请务必保证 ca-key.pem 文件的安全,*.csr 文件在整个过程中不会使用。
4,生成服务器端证书
- $ echo '{"CN":"coreos1","hosts":["10.93.81.17","127.0.0.1"],"key":{"algo":"rsa","size":2048}}' | cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=server -hostname="10.93.81.17,127.0.0.1,server" - | cfssljson -bare server
- hosts 字段需要自定义。
- 然后将得到以下几个文件:
- server-key.pem
- server.csr
- server.pem
5,生成对等证书
- $ echo '{"CN":"member1","hosts":["10.93.81.17","127.0.0.1"],"key":{"algo":"rsa","size":2048}}' | cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=peer -hostname="10.93.81.17,127.0.0.1,server,member1" - | cfssljson -bare member1
- hosts 字段需要自定义。
- 然后将得到以下几个文件:
- member1-key.pem
- member1.csr
- member1.pem
- 如果有多个 etcd 成员,重复此步为每个成员生成对等证书。
6,生成客户端证书
- $ echo '{"CN":"client","hosts":["10.93.81.17","127.0.0.1"],"key":{"algo":"rsa","size":2048}}' | cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=client - | cfssljson -bare client
- hosts 字段需要自定义。
- 然后将得到以下几个文件:
- client-key.pem
- client.csr
- client.pem
至此,所有证书都已生成完毕。
2,拷贝密钥对到所有节点
- 1,拷贝密钥对到所有节点
- 2,更新系统证书库
1,拷贝密钥对到所有节点
- $ mkdir -pv /etc/ssl/etcd/
- $ cp ~/cfssl/* /etc/ssl/etcd/
- $ chown -R etcd:etcd /etc/ssl/etcd
- $ chmod 600 /etc/ssl/etcd/*-key.pem
- $ cp ~/cfssl/ca.pem /etc/ssl/certs/
2,更新系统证书库
- $ yum install ca-certificates -y
- $ update-ca-trust
3,配置 etcd 使用证书
- $ etcdctl version
- etcdctl version: 3.1.3
- API version: 3.1
- $ cat /etc/etcd/etcd.conf
- ETCD_DATA_DIR="/var/lib/etcd/default.etcd"
- #监听URL,用于与其他节点通讯
- ETCD_LISTEN_PEER_URLS="https://10.93.81.17:2380"
- #告知客户端的URL, 也就是服务的URL
- ETCD_LISTEN_CLIENT_URLS="https://10.93.81.17:2379,https://10.93.81.17:4001"
- #表示监听其他节点同步信号的地址
- ETCD_INITIAL_ADVERTISE_PEER_URLS="https://10.93.81.17:2380"
- #–advertise-client-urls 告知客户端的URL, 也就是服务的URL,tcp2379端口用于监听客户端请求
- ETCD_ADVERTISE_CLIENT_URLS="https://10.93.81.17:2379"
- #启动参数配置
- ETCD_NAME="node1"
- ETCD_INITIAL_CLUSTER="node1=https://10.93.81.17:2380"
- ETCD_INITIAL_CLUSTER_STATE="new"
- #[security]
- ETCD_CERT_FILE="/etc/ssl/etcd/server.pem"
- ETCD_KEY_FILE="/etc/ssl/etcd/server-key.pem"
- ETCD_TRUSTED_CA_FILE="/etc/ssl/etcd/ca.pem"
- ETCD_CLIENT_CERT_AUTH="true"
- ETCD_PEER_CERT_FILE="/etc/ssl/etcd/member1.pem"
- ETCD_PEER_KEY_FILE="/etc/ssl/etcd/member1-key.pem"
- ETCD_PEER_TRUSTED_CA_FILE="/etc/ssl/etcd/ca.pem"
- ETCD_PEER_CLIENT_CERT_AUTH="true"
- #[logging]
- ETCD_DEBUG="true"
- ETCD_LOG_PACKAGE_LEVELS="etcdserver=WARNING,security=DEBUG"
4,测试 etcd 是否正常
- $ systemctl restart etcd
- 如果报错,使用 journalctl -f -t etcd 和 journalctl -u etcd 来定位问题。
- $ curl --cacert /etc/ssl/etcd/ca.pem --cert /etc/ssl/etcd/client.pem --key /etc/ssl/etcd/client-key.pem https://10.93.81.17:2379/health
- {"health": "true"}
- $ etcdctl --endpoints=[10.93.81.17:2379] --cacert=/etc/ssl/etcd/ca.pem --cert=/etc/ssl/etcd/client.pem --key=/etc/ssl/etcd/client-key.pem member list
- $ etcdctl --endpoints=[10.93.81.17:2379] --cacert=/etc/ssl/etcd/ca.pem --cert=/etc/ssl/etcd/client.pem --key=/etc/ssl/etcd/client-key.pem put /foo/bar "hello world"
- $ etcdctl --endpoints=[10.93.81.17:2379] --cacert=/etc/ssl/etcd/ca.pem --cert=/etc/ssl/etcd/client.pem --key=/etc/ssl/etcd/client-key.pem get /foo/bar
5,配置 kube-apiserver 使用 CA 连接 etcd
- $ cp /etc/ssl/etcd/* /var/run/kubernetes/
- $ chown -R kube.kube /var/run/kubernetes/
在 /etc/kubernetes/apiserver 中 KUBE_API_ARGS 新加一下几个参数:
- --cert-dir='/var/run/kubernetes/' --etcd-cafile='/var/run/kubernetes/ca.pem' --etcd-certfile='/var/run/kubernetes/client.pem' --etcd-keyfile='/var/run/kubernetes/client-key.pem'
6,测试 kube-apiserver
- $ systemctl restart kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy
- $ systemctl status -l kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy
- $ kubectl get node
- $ kubectl get cs
- NAME STATUS MESSAGE ERROR
- scheduler Healthy ok
- controller-manager Healthy ok
- etcd-0 Unhealthy Get https://10.93.81.17:2379/health: remote error: tls: bad certificate
- $ ./version.sh
- etcdctl version: 3.1.3
- API version: 3.1
- Kubernetes v1.6.0-beta.1
7,未解决的问题
1,使用 kubectl get cs
查看会出现如上面所示的报错:
- etcd-0 Unhealthy Get https://10.93.81.17:2379/health: remote error: tls: bad certificate
此问题有人提交 pr 但尚未被 merge,etcd component status check should include credentials
2,使用以下命令查看到的 2380 端口是未加密的
- $ etcdctl --endpoints=[10.93.81.17:2379] --cacert=/etc/ssl/etcd/ca.pem --cert=/etc/ssl/etcd/client.pem --key=/etc/ssl/etcd/client-key.pem member list
- 2017-03-15 15:02:05.611564 I | warning: ignoring ServerName for user-provided CA for backwards compatibility is deprecated
- 145b401ad8709f51, started, node1, http://10.93.81.17:2380, https://10.93.81.17:2379
参考文档:
来源:https://www.cnblogs.com/fungitive/p/9136196.html