场景一: 将新版本灰度给部分用户
假设线上运行了一套对外提供 7 层服务的 Service A 服务,后来开发了个新版本 Service A’ 想 要上线,但又不想直接替换掉原来的 Service A,希望先灰度一小部分用户,等运行一段时间足够稳定 了再逐渐全量上线新版本,最后平滑下线旧版本。这个时候就可以利用 Nginx Ingress 基于 Header 或 Cookie 进行流量切分的策略来发布,业务使用 Header 或 Cookie 来标识不同类型的用户,我们通过配置 Ingress 来实现让带有指定 Header 或 Cookie 的请求被转发到新版本,其它的仍然转发到旧版本,从而实现将新版本灰度给部分用户:
场景二: 切一定比例的流量给新版本
假设线上运行了一套对外提供 7 层服务的 Service B 服务,后来修复了一些问题,需要灰度上线 一个新版本 Service B’,但又不想直接替换掉原来的 Service B,而是让先切 10% 的流量到新版 本,等观察一段时间稳定后再逐渐加大新版本的流量比例直至完全替换旧版本,最后再滑下线旧版本,从 而实现切一定比例的流量给新版本:
Ingress-Nginx 是一个 K8S ingress 工具,支持配置 Ingress Annotations 来实现不同场景下的灰度发布和测试。 Nginx Annotations 支持以下几种 Canary 规则:
假设我们现在部署了两个版本的服务,老版本和 canary 版本
nginx.ingress.kubernetes.io/canary-by-header:基于 Request Header 的流量切分,适用于 灰度发布以及 A/B 测试。当 Request Header 设置为 always 时,请求将会被一直发送到 Canary 版 本;当 Request Header 设置为 never 时,请求不会被发送到 Canary 入口。
nginx.ingress.kubernetes.io/canary-by-header-value:要匹配的 Request Header 的值, 用于通知 Ingress 将请求路由到 Canary Ingress 中指定的服务。当 Request Header 设置为此值 时,它将被路由到 Canary 入口。
nginx.ingress.kubernetes.io/canary-weight:基于服务权重的流量切分,适用于蓝绿部署,权重范围 0 - 100 按百分比将请求路由到 Canary Ingress 中指定的服务。权重为 0 意味着该金丝雀规则不会向 Canary 入口的服务发送任何请求。权重为 60 意味着 60%流量转到 canary。权重为 100 意 味着所有请求都将被发送到 Canary 入口。
nginx.ingress.kubernetes.io/canary-by-cookie:基于Cookie 的流量切分,适用于灰度发布 与 A/B 测试。用于通知 Ingress 将请求路由到 Canary Ingress 中指定的服务的 cookie。当cookie 值设置为 always 时,它将被路由到 Canary 入口;当 cookie 值设置为 never 时,请求不会 被发送到 Canary 入口。
# 1.node1和node3节点导入镜像 docker load -i openresty.tar.gz Loaded image: openresty/openresty:centos # 2.部署v1版本 [root@master header-ingress]# cat v1.yaml apiVersion: apps/v1 kind: Deployment metadata: name: nginx-v1 spec: replicas: 1 selector: matchLabels: app: nginx version: v1 template: metadata: labels: app: nginx version: v1 spec: containers: - name: nginx image: "openresty/openresty:centos" imagePullPolicy: IfNotPresent ports: - name: http protocol: TCP containerPort: 80 volumeMounts: - mountPath: /usr/local/openresty/nginx/conf/nginx.conf name: config subPath: nginx.conf volumes: - name: config configMap: name: nginx-v1 --- apiVersion: v1 kind: ConfigMap metadata: labels: app: nginx version: v1 name: nginx-v1 data: nginx.conf: |- worker_processes 1; events { accept_mutex on; multi_accept on; use epoll; worker_connections 1024; } http { ignore_invalid_headers off; server { listen 80; location / { access_by_lua ' local header_str = ngx.say("nginx-v1") '; } } } --- apiVersion: v1 kind: Service metadata: name: nginx-v1 spec: type: ClusterIP ports: - port: 80 protocol: TCP name: http selector: app: nginx version: v1 [root@master header-ingress]# kubectl apply -f v1.yaml deployment.apps/nginx-v1 created configmap/nginx-v1 created service/nginx-v1 created [root@master header-ingress]# kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE nginx-v1 ClusterIP 10.108.228.242 <none> 80/TCP 31s tomcat ClusterIP 10.109.238.78 <none> 8080/TCP,8009/TCP 3h20m [root@master header-ingress]# kubectl get configmap NAME DATA AGE nginx-v1 1 51s [root@master header-ingress]# kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE nginx-v1 0/1 0 0 78s [root@master header-ingress]# kubectl get pods NAME READY STATUS RESTARTS AGE nginx-v1-79bc94ff97-7bfg4 1/1 Running 0 55s # 3.部署v2版本 [root@master header-ingress]# cat v2.yaml apiVersion: apps/v1 kind: Deployment metadata: name: nginx-v2 spec: replicas: 1 selector: matchLabels: app: nginx version: v2 template: metadata: labels: app: nginx version: v2 spec: containers: - name: nginx image: "openresty/openresty:centos" imagePullPolicy: IfNotPresent ports: - name: http protocol: TCP containerPort: 80 volumeMounts: - mountPath: /usr/local/openresty/nginx/conf/nginx.conf name: config subPath: nginx.conf volumes: - name: config configMap: name: nginx-v2 --- apiVersion: v1 kind: ConfigMap metadata: labels: app: nginx version: v2 name: nginx-v2 data: nginx.conf: |- worker_processes 1; events { accept_mutex on; multi_accept on; use epoll; worker_connections 1024; } http { ignore_invalid_headers off; server { listen 80; location / { access_by_lua ' local header_str = ngx.say("nginx-v2") '; } } } --- apiVersion: v1 kind: Service metadata: name: nginx-v2 spec: type: ClusterIP ports: - port: 80 protocol: TCP name: http selector: app: nginx version: v2 [root@master header-ingress]# kubectl get pods NAME READY STATUS RESTARTS AGE nginx-v1-79bc94ff97-7bfg4 1/1 Running 0 3m21s nginx-v2-5f885975d5-4pnkq 1/1 Running 0 4s [root@master header-ingress]# kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE nginx-v1 ClusterIP 10.108.228.242 <none> 80/TCP 7m13s nginx-v2 ClusterIP 10.104.97.248 <none> 80/TCP 10s # 4.再创建一个Ingress,对外暴露服务,指向v1版本服务 apiVersion: extensions/v1beta1 kind: Ingress metadata: name: nginx annotations: kubernetes.io/ingress.class: nginx spec: rules: - host: canary.example.com http: paths: - backend: serviceName: nginx-v1 servicePort: 80 path: / [root@master header-ingress]# kubectl apply -f v1-ingress.yaml Warning: extensions/v1beta1 Ingress is deprecated in v1.14+, unavailable in v1.22+; use networking.k8s.io/v1 Ingress ingress.extensions/nginx created [root@master header-ingress]# kubectl get ingress NAME CLASS HOSTS ADDRESS PORTS AGE nginx <none> canary.example.com 192.168.10.11,192.168.10.13 80 30s #访问验证结果 curl -H "Host: canary.example.com" http://EXTERNAL-IP [root@master header-ingress]# curl -H "Host: canary.example.com" http://192.168.10.11 nginx-v1 # EXTERNAL-IP替换为nginx Ingress自身对外暴露的IP
创建 Canary Ingress,指定 v2 版本的后端服务,且加上一些 annotation,实现仅将带有名为 Region 且值为 cd 或 sz 的请求头的请求转发给当前 Canary Ingress,模拟灰度新版本给成都和深 圳地域的用户
[root@master header-ingress]# cat v2-ingress.yaml apiVersion: extensions/v1beta1 kind: Ingress metadata: annotations: kubernetes.io/ingress.class: nginx nginx.ingress.kubernetes.io/canary: "true" nginx.ingress.kubernetes.io/canary-by-header: "Region" nginx.ingress.kubernetes.io/canary-by-header-pattern: "cd|sz" name: nginx-canary spec: rules: - host: canary.example.com http: paths: - backend: serviceName: nginx-v2 servicePort: 80 path: / [root@master header-ingress]# kubectl apply -f v2-ingress.yaml Warning: extensions/v1beta1 Ingress is deprecated in v1.14+, unavailable in v1.22+; use networking.k8s.io/v1 Ingress ingress.extensions/nginx-canary created [root@master header-ingress]# kubectl get ingress NAME CLASS HOSTS ADDRESS PORTS AGE nginx <none> canary.example.com 192.168.10.11,192.168.10.13 80 11m nginx-canary <none> canary.example.com 192.168.10.11,192.168.10.13 80 47s 测试访问: curl -H "Host: canary.example.com" -H "Region: cd" http://EXTERNAL-IP # EXTERNAL-IP 替换为 Nginx Ingress 自身对外暴露的 IP [root@master header-ingress]# curl -H "Host:canary.example.com" -H "Region:cd" http://192.168.10.11 nginx-v2 [root@master header-ingress]# curl -H "Host:canary.example.com" -H "Region:sh" http://192.168.10.11 nginx-v1 [root@master header-ingress]# curl -H "Host:canary.example.com" -H "Region:cd" http://192.168.10.199 nginx-v2
[root@master header-ingress]# curl -H "Host:canary.example.com" -H "Region:sz" http://192.168.10.199
nginx-v2
[root@master header-ingress]# curl -H "host:canary.example.com" http://192.168.10.11 nginx-v1 可以看到,只有 header Region 为 cd 或 sz 的请求才由 v2 版本服务响应。
与前面 Header 类似,不过使用 Cookie 就无法自定义 value 了,这里以模拟灰度成都地域用户 为例,仅将带有名为 user_from_cd 的 cookie 的请求转发给当前 Canary Ingress 。先删除前面基 于 Header 的流量切分的 Canary Ingress,然后创建下面新的 Canary Ingress:
[root@master header-ingress]# kubectl delete -f v2-ingress.yaml [root@master header-ingress]# cat v1-cookie.yaml apiVersion: extensions/v1beta1 kind: Ingress metadata: annotations: kubernetes.io/ingress.class: nginx nginx.ingress.kubernetes.io/canary: "true" nginx.ingress.kubernetes.io/canary-by-cookie: "user_from_cd" name: nginx-canary spec: rules: - host: canary.example.com http: paths: - backend: serviceName: nginx-v2 servicePort: 80 path: / [root@master header-ingress]# kubectl apply -f v1-cookie.yaml Warning: extensions/v1beta1 Ingress is deprecated in v1.14+, unavailable in v1.22+; use networking.k8s.io/v1 Ingress ingress.extensions/nginx-canary created [root@master header-ingress]# kubectl get ingress NAME CLASS HOSTS ADDRESS PORTS AGE nginx-canary <none> canary.example.com 192.168.10.11,192.168.10.13 80 30s 测试访问: [root@master header-ingress]# curl -s -H "Host:canary.example.com" --cookie "user_from_cd=always" http://192.168.10.199 nginx-v2 [root@master header-ingress]# curl -s -H "Host:canary.example.com" --cookie "user_from_sh=always" http://192.168.10.199 nginx-v1 [root@master header-ingress]# curl -s -H "Host:canary.example.com" http://192.168.10.199 nginx-v1 可以看到,只有 cookie user_from_cd 为 always 的请求才由 v2 版本的服务响应。
基于服务权重的 Canary Ingress 就简单了,直接定义需要导入的流量比例,这里以导入 10% 流 量到 v2 版本为例 (如果有,先删除之前的 Canary Ingress):
[root@master header-ingress]# kubectl delete -f v1-cookie.yaml [root@master header-ingress]# cat v1-weight.yaml apiVersion: extensions/v1beta1 kind: Ingress metadata: annotations: kubernetes.io/ingress.class: nginx nginx.ingress.kubernetes.io/canary: "true" nginx.ingress.kubernetes.io/canary-weight: "10" name: nginx-canary spec: rules: - host: canary.example.com http: paths: - backend: serviceName: nginx-v2 servicePort: 80 path: / [root@master header-ingress]# kubectl apply -f v1-weight.yaml Warning: extensions/v1beta1 Ingress is deprecated in v1.14+, unavailable in v1.22+; use networking.k8s.io/v1 Ingress ingress.extensions/nginx-canary created [root@master header-ingress]# kubectl get ingress NAME CLASS HOSTS ADDRESS PORTS AGE nginx <none> canary.example.com 192.168.10.11,192.168.10.13 80 28m nginx-canary <none> canary.example.com 80 13s [root@master header-ingress]# for i in {1..10}; do curl -H "Host: canary.example.com" http://192.168.10.199;done; nginx-v1 nginx-v2 nginx-v1 nginx-v1 nginx-v1 nginx-v1 nginx-v1 nginx-v1 nginx-v2 nginx-v1 可以看到,大概只有十分之一的几率由 v2 版本的服务响应,符合 10% 服务权重的设置,目前只有10个所以看到有2个v2