欢迎光临
我的个人博客网站

容器编排系统K8s之NetworkPolicy资源

  前文我们了解了k8s的网络插件flannel的基础工作逻辑,回顾请参考:https://www.cnblogs.com/qiuhom-1874/p/14225657.html;今天我们来聊一下k8s上的NetworkPolicy相关话题;

  NetworkPolicy资源是做什么用的?

  我们知道在k8s上我们可以用名称空间来隔离多个资源,在不同名称空间下我们可以创建相同名称和相同类型的资源;有些资源还必须依赖名称空间才可以被创建;但是在k8s上,名称空间不可以隔离网络,所谓隔离网络是指不管pod创建在那个名称空间下,都可以被其他名称空间下pod访问到;如果我们想要限制对应名称空间下的pod被别的名称空间或外部客户端访问,我们可以在对应名称空间下创建NetworkPolicy资源;该资源可以限制哪些pod可以被外部客户端访问,或者哪些pod不能够被哪个或哪些名称空间的pod访问;除了能够限制对应名称空间下的pod网络入站流量,它还可以限制对应名称空间下的pod出站的流量,允许或拒绝某个或某些个pod访问外部某名称空间下pod或外部某服务等;我们可以简单的理解NetworkingPolicy就是对应名称空间的一道防火墙;我们把NetworkPolicy资源创建在那个名称空间下,那么它就能实现对该名称空间下的pod进行流量管控;

  NetworkPolicy基础工作逻辑

容器编排系统K8s之NetworkPolicy资源

  提示:首先networkpolicy是作用于一个名称空间下的pod资源之上而进行的流量管控;它可以通过pod选择器来选择对应名称空间下pod,满足对应选择器的条件的pod,将受到networkpolicy中定义的策略管控,默认pod选择器没有明确定义表示匹配对应名称空间下的所有pod;Egress是用来定义pod出站策略,默认显式定义的策略都是被允许出站的,没有定义表示拒绝;对于ingress也是同样的逻辑,只有显式定义的规则才被允许,没有定义就表示拒绝;对于出站流量来说,默认我们定义或不定义它都不会生效,都是允许放行的,只有显式配置对应出站规则生效,它才会生效,对应定义的出站流量才会被限制;

  示例:定义dev名称空间下的pod不允许任何客户端访问

[root@master01 ~]# cat denyall-ingress-dev.yaml apiVersion: networking.k8s.io/v1 kind: NetworkPolicy metadata:   name: deny-all-ingress   namespace: dev spec:   podSelector: {}   policyTypes:   - Ingress [root@master01 ~]#  

  提示:以上资源清单表示对dev名称空间下的pod的入站流量进行管控,拒绝所有客户端访问对应dev名称空间下的pod;这其中包含dev名称空间下的pod;

  应用资源清单

  在应用之前请确保你说部署的网络插件支持网络策略,默认flannel网络插件不支持网络策略,我们需要部署canel这个网络插件,该插件能够结合flannel插件,让其拥有calico的网络策略功能;

  部署canel

[root@master01 ~]# kubectl apply -f https://docs.projectcalico.org/manifests/canal.yaml configmap/canal-config created customresourcedefinition.apiextensions.k8s.io/bgpconfigurations.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/bgppeers.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/blockaffinities.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/clusterinformations.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/felixconfigurations.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/globalnetworkpolicies.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/globalnetworksets.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/hostendpoints.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/ipamblocks.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/ipamconfigs.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/ipamhandles.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/ippools.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/kubecontrollersconfigurations.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/networkpolicies.crd.projectcalico.org created customresourcedefinition.apiextensions.k8s.io/networksets.crd.projectcalico.org created clusterrole.rbac.authorization.k8s.io/calico-kube-controllers created clusterrolebinding.rbac.authorization.k8s.io/calico-kube-controllers created clusterrole.rbac.authorization.k8s.io/calico-node created clusterrole.rbac.authorization.k8s.io/flannel configured clusterrolebinding.rbac.authorization.k8s.io/canal-flannel created clusterrolebinding.rbac.authorization.k8s.io/canal-calico created daemonset.apps/canal created serviceaccount/canal created deployment.apps/calico-kube-controllers created serviceaccount/calico-kube-controllers created poddisruptionbudget.policy/calico-kube-controllers created [root@master01 ~]#  

  提示:等待上述的pod都正常运行以后,我们就可以应用上述清单,让其网络策略生效;

  创建dev名称空间,并在其下创建一个deploy控制器,控制器创建3个nginx pod

[root@master01 manifests]# cat pod-demo.yaml apiVersion: v1 kind: Namespace metadata:   name: dev   labels:     name: dev --- apiVersion: apps/v1 kind: Deployment metadata:   name: nginx-dep   namespace: dev spec:   replicas: 3   selector:     matchLabels:       app: nginx       rel: stable   template:     metadata:       labels:         app: nginx         rel: stable     spec:       containers:       - name: nginx         image: nginx:1.14-alpine         imagePullPolicy: IfNotPresent         ports:         - name: http           containerPort: 80 [root@master01 manifests]# kubectl apply -f pod-demo.yaml namespace/dev unchanged deployment.apps/nginx-dep created [root@master01 manifests]# kubectl get pods -n dev -o wide NAME                        READY   STATUS    RESTARTS   AGE   IP           NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-5d9tb   1/1     Running   0          51s   10.244.4.6   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-ftmn9   1/1     Running   0          51s   10.244.1.5   node01.k8s.org   <none>           <none> nginx-dep-cb74c595f-jf9p9   1/1     Running   0          51s   10.244.3.4   node03.k8s.org   <none>           <none> [root@master01 manifests]#  

  在没有应用上述网络策略前,用default里的pod访问对应dev名称空间下pod,看看是否可以正常访问?

[root@master01 manifests]# kubectl get pods -n dev -o wide NAME                        READY   STATUS    RESTARTS   AGE   IP           NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-5d9tb   1/1     Running   0          86s   10.244.4.6   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-ftmn9   1/1     Running   0          86s   10.244.1.5   node01.k8s.org   <none>           <none> nginx-dep-cb74c595f-jf9p9   1/1     Running   0          86s   10.244.3.4   node03.k8s.org   <none>           <none> [root@master01 manifests]# kubectl get pods  NAME    READY   STATUS    RESTARTS   AGE web-0   1/1     Running   6          3d16h web-1   1/1     Running   5          3d16h web-2   1/1     Running   6          3d16h [root@master01 manifests]# kubectl exec web-0 -it -- /bin/sh / # wget -O - -q 10.244.4.6 <!DOCTYPE html> <html> <head> <title>Welcome to nginx!</title> <style>     body {         width: 35em;         margin: 0 auto;         font-family: Tahoma, Verdana, Arial, sans-serif;     } </style> </head> <body> <h1>Welcome to nginx!</h1> <p>If you see this page, the nginx web server is successfully installed and working. Further configuration is required.</p>  <p>For online documentation and support please refer to <a href="http://nginx.org/">nginx.org</a>.<br/> Commercial support is available at <a href="http://nginx.com/">nginx.com</a>.</p>  <p><em>Thank you for using nginx.</em></p> </body> </html> / #  

  提示:可以看到用default名称空间下的pod,是可以正常访问dev名称空间下的pod;

  应用资源网络策略清单

[root@master01 ~]# cat denyall-ingress-dev.yaml  apiVersion: networking.k8s.io/v1 kind: NetworkPolicy metadata:   name: deny-all-ingress   namespace: dev spec:   podSelector: {}   policyTypes:   - Ingress [root@master01 ~]# kubectl apply -f denyall-ingress-dev.yaml networkpolicy.networking.k8s.io/deny-all-ingress created [root@master01 ~]# kubectl get netpol -n dev NAME               POD-SELECTOR   AGE deny-all-ingress   <none>         10s [root@master01 ~]# kubectl describe netpol deny-all-ingress -n dev Name:         deny-all-ingress Namespace:    dev Created on:   2021-01-04 15:34:40 +0800 CST Labels:       <none> Annotations:  <none> Spec:   PodSelector:     <none> (Allowing the specific traffic to all pods in this namespace)   Allowing ingress traffic:     <none> (Selected pods are isolated for ingress connectivity)   Not affecting egress traffic   Policy Types: Ingress [root@master01 ~]#  

  提示:从上面的netpol资源的详细信息中可以了解到,对应netpol资源pod选择器为none表示选择对应名称空间下的所有pod,allowing ingress traffic 也是为none表示没有任何客户端允许访问该名称空间下的pod;没有出站策略,表示拒绝所有出站流量;默认只是入站规则生效,所以这里的出站规则不生效,即不限制出站流量;policyTypes表示对应的那个策略生效,默认不指定是Ingress;

  验证:使用default名称空间下的pod访问dev名称空间下的pod,看看是否还可以正常访问?

容器编排系统K8s之NetworkPolicy资源

  提示:可以看到现在default名称空间下的pod是不能访问dev名称空间下的pod;

  验证:dev名称空间下的pod访问同名称空间下的pod,看看是否可以允许访问?

容器编排系统K8s之NetworkPolicy资源

  提示:可以看到在dev名称空间下的pod也没法访问dev名称空间下的pod,但是自己可以访问自己,其原因是自己访问自己,流量不会出站,在lo接口就被处理了,所以自己访问自己不受限制;

  验证:使用dev名称空间下的pod访问default名称空间下的pod,看看是否被允许 ?

容器编排系统K8s之NetworkPolicy资源

  提示:可以看到使用dev名称空间下的pod访问default名称空间下的pod上没有问题,说明刚才的那条规则只是现在了dev名称空间下的pod的入站流量,并没有限制其出站流量;上面访问10.244.4.5和3.3响应404,是因为对应pod上的web服务没有主页;

  示例:限制dev名称空间下的pod的出站流量,拒绝所有pod流量出站

[root@master01 ~]# cat denyall-Egress.yaml apiVersion: networking.k8s.io/v1 kind: NetworkPolicy metadata:    name: deny-all-egress   namespace: dev spec:   podSelector: {}   policyTypes:   - Egress [root@master01 ~]#  

  提示:policyTypes字段用来控制那个类型的规则生效,该字段是一个列表,默认是Ingress,表示只对入站流量做限制,如果明确手动定义为Egress表示,只限制出站流量;出站策略和入站策略的逻辑都是一样的,只要没有明且定义的,都是表示拒绝操作,只有明确定义的才表示对应允许操作;上述没有定义任何出站规则,表示拒绝所有pod访问外部服务;

  应用资源清单

[root@master01 ~]# kubectl apply -f denyall-Egress.yaml networkpolicy.networking.k8s.io/deny-all-egress created [root@master01 ~]# kubectl get netpol -n dev NAME               POD-SELECTOR   AGE deny-all-egress    <none>         10s deny-all-ingress   <none>         24m [root@master01 ~]# kubectl describe netpol deny-all-egress -n dev Name:         deny-all-egress Namespace:    dev Created on:   2021-01-04 15:59:16 +0800 CST Labels:       <none> Annotations:  <none> Spec:   PodSelector:     <none> (Allowing the specific traffic to all pods in this namespace)   Not affecting ingress traffic   Allowing egress traffic:     <none> (Selected pods are isolated for egress connectivity)   Policy Types: Egress [root@master01 ~]#  

  验证:进入dev名称空间下的pod交互式接口,看看是否可以访问其他名称空间下的pod?

容器编排系统K8s之NetworkPolicy资源

  提示:可以看到现在dev名称空间下的pod访问default名称空间下的pod就不能访问了;

  验证:删除入站限制,使用dev名称空间下的pod互相访问,看看是否可以正常访问呢?

[root@master01 ~]# kubectl get netpol -n dev NAME               POD-SELECTOR   AGE deny-all-egress    <none>         5m50s deny-all-ingress   <none>         30m [root@master01 ~]# kubectl delete netpol deny-all-ingress -n dev networkpolicy.networking.k8s.io "deny-all-ingress" deleted [root@master01 ~]# kubectl get netpol -n dev                     NAME              POD-SELECTOR   AGE deny-all-egress   <none>         6m9s [root@master01 ~]# kubectl get pods -n dev -o wide                              NAME                        READY   STATUS    RESTARTS   AGE   IP           NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-5d9tb   1/1     Running   0          34m   10.244.4.6   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-ftmn9   1/1     Running   0          34m   10.244.1.5   node01.k8s.org   <none>           <none> nginx-dep-cb74c595f-jf9p9   1/1     Running   0          34m   10.244.3.4   node03.k8s.org   <none>           <none> [root@master01 ~]# kubectl exec -it nginx-dep-cb74c595f-5d9tb -n dev -- /bin/sh / # wget -O - -q 10.244.1.5 ^C / # wget -O - -q 10.244.3.4 ^C / # wget -O - -q 10.244.4.6 <!DOCTYPE html> <html> <head> <title>Welcome to nginx!</title> <style>     body {         width: 35em;         margin: 0 auto;         font-family: Tahoma, Verdana, Arial, sans-serif;     } </style> </head> <body> <h1>Welcome to nginx!</h1> <p>If you see this page, the nginx web server is successfully installed and working. Further configuration is required.</p>  <p>For online documentation and support please refer to <a href="http://nginx.org/">nginx.org</a>.<br/> Commercial support is available at <a href="http://nginx.com/">nginx.com</a>.</p>  <p><em>Thank you for using nginx.</em></p> </body> </html> / #  

  提示:可以看到限制了出站流量以后,在同一名称空间下的pod也是无法正常访问的;但是自己可以访问自己;

  示例:限制dev名称空间下的pod的出站流量,允许dev名称空间下的pod互相访问

[root@master01 ~]# cat allow-dev-Egress.yaml apiVersion: networking.k8s.io/v1 kind: NetworkPolicy metadata:   name: allow-dev-egress   namespace: dev spec:   podSelector: {}   egress:     - to:       - podSelector:           matchLabels:             app: nginx             rel: stable       ports:       - port: 80         protocol: TCP    policyTypes:   - Egress [root@master01 ~]#  

  提示:限制出站流量,可以在spec.egress中来定义允许的出站的规则;其中to字段是用来描述被访问端,其值类型是一个列表对象;podSelector字段是用来描述被访问端pod选择器,只有满足对应pod选择器的条件的pod都被允许出站;spec.egress.ports字段是用来描述被访问端点端口信息;该字段为一个列表对象;这里需要注意,默认没有在规则指定其名称空间表示匹配当前netpol所在名称空间;以上清单表示dev下的pod允许访问dev名称空间下标签为app=nginx,rel=stable的pod的80端口;

  删除deny-all-egress出站策略,应用其配置清单

[root@master01 ~]# kubectl get netpol -n dev NAME              POD-SELECTOR   AGE deny-all-egress   <none>         26m [root@master01 ~]# kubectl delete netpol deny-all-egress -n dev networkpolicy.networking.k8s.io "deny-all-egress" deleted [root@master01 ~]# kubectl apply -f allow-dev-Egress.yaml networkpolicy.networking.k8s.io/allow-dev-egress created [root@master01 ~]# kubectl get netpol -n dev NAME               POD-SELECTOR   AGE allow-dev-egress   <none>         10s [root@master01 ~]# kubectl describe netpol allow-dev-egress -n dev Name:         allow-dev-egress Namespace:    dev Created on:   2021-01-04 16:34:23 +0800 CST Labels:       <none> Annotations:  <none> Spec:   PodSelector:     <none> (Allowing the specific traffic to all pods in this namespace)   Not affecting ingress traffic   Allowing egress traffic:     To Port: 80/TCP     To:       PodSelector: app=nginx,rel=stable   Policy Types: Egress [root@master01 ~]#  

  验证:进入dev名称空间下pod的里,看看是否可以正常访问dev名称空间下的pod?

[root@master01 ~]# kubectl get pods -n dev -o wide NAME                        READY   STATUS    RESTARTS   AGE   IP           NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-5d9tb   1/1     Running   0          65m   10.244.4.6   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-ftmn9   1/1     Running   0          65m   10.244.1.5   node01.k8s.org   <none>           <none> nginx-dep-cb74c595f-jf9p9   1/1     Running   0          65m   10.244.3.4   node03.k8s.org   <none>           <none> [root@master01 ~]# kubectl exec -it -n dev nginx-dep-cb74c595f-5d9tb -- /bin/sh / # wget -O - -q 10.244.1.5 <!DOCTYPE html> <html> <head> <title>Welcome to nginx!</title> <style>     body {         width: 35em;         margin: 0 auto;         font-family: Tahoma, Verdana, Arial, sans-serif;     } </style> </head> <body> <h1>Welcome to nginx!</h1> <p>If you see this page, the nginx web server is successfully installed and working. Further configuration is required.</p>  <p>For online documentation and support please refer to <a href="http://nginx.org/">nginx.org</a>.<br/> Commercial support is available at <a href="http://nginx.com/">nginx.com</a>.</p>  <p><em>Thank you for using nginx.</em></p> </body> </html> / # wget -O - -q 10.244.3.4 <!DOCTYPE html> <html> <head> <title>Welcome to nginx!</title> <style>     body {         width: 35em;         margin: 0 auto;         font-family: Tahoma, Verdana, Arial, sans-serif;     } </style> </head> <body> <h1>Welcome to nginx!</h1> <p>If you see this page, the nginx web server is successfully installed and working. Further configuration is required.</p>  <p>For online documentation and support please refer to <a href="http://nginx.org/">nginx.org</a>.<br/> Commercial support is available at <a href="http://nginx.com/">nginx.com</a>.</p>  <p><em>Thank you for using nginx.</em></p> </body> </html> / #  

  提示:可以看到dev名称空间pod访问dev名称空间下的pod可以正常访问;

  验证:使用dev名称空间pod访问default名称空间下,标签为app=nginx,rel=stable的pod,看看是否可以正常访问?

[root@master01 ~]# kubectl get pods -n dev  NAME                        READY   STATUS    RESTARTS   AGE nginx-dep-cb74c595f-27qjp   1/1     Running   2          88m nginx-dep-cb74c595f-b92s4   1/1     Running   1          88m nginx-dep-cb74c595f-wdqnh   1/1     Running   1          88m [root@master01 ~]# kubectl get pods -o wide --show-labels                       NAME                        READY   STATUS    RESTARTS   AGE     IP            NODE             NOMINATED NODE   READINESS GATES   LABELS nginx-dep-cb74c595f-5mwl2   1/1     Running   1          89m     10.244.4.18   node04.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-vxc9z   1/1     Running   1          89m     10.244.3.12   node03.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-z265t   1/1     Running   1          89m     10.244.2.13   node02.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable web-0                       1/1     Running   8          3d20h   10.244.2.14   node02.k8s.org   <none>           <none>            app=nginx,controller-revision-hash=web-6db9455fdf,statefulset.kubernetes.io/pod-name=web-0 web-1                       1/1     Running   7          3d20h   10.244.4.16   node04.k8s.org   <none>           <none>            app=nginx,controller-revision-hash=web-6db9455fdf,statefulset.kubernetes.io/pod-name=web-1 web-2                       1/1     Running   8          3d20h   10.244.3.11   node03.k8s.org   <none>           <none>            app=nginx,controller-revision-hash=web-6db9455fdf,statefulset.kubernetes.io/pod-name=web-2 [root@master01 ~]# kubectl exec -it nginx-dep-cb74c595f-27qjp -n dev -- /bin/sh / # wget -O - -q 10.244.4.18 ^C / # wget -O - -q 10.244.3.12 ^C / # wget -O - -q 10.244.2.13 ^C / #  

  提示:可以看到dev名称空间下的pod访问default名称空间下标签为app=nginx,rel=stable的pod也是无法正常访问,原因是我们在策略中没有指定对应访问目标的名称空间,默认就是对应netpol所在名称空间;

  示例:允许dev名称空间下的pod访问default和dev名称空间下标签为app=nginx,rel=stable的pod

[root@master01 ~]# cat allow-dev-def-egress.yaml apiVersion: networking.k8s.io/v1 kind: NetworkPolicy metadata:   name: allow-dev-egress   namespace: dev spec:   podSelector: {}    egress:     - to:       - namespaceSelector:           matchExpressions:           - key: name             operator: In             values: ["def","dev"]         podSelector:           matchLabels:             app: nginx             rel: stable       ports:       - port: 80         protocol: TCP    policyTypes:   - Egress [root@master01 ~]#  

  提示:spec.egress.to中的namespeceSelector字段使用来描述被访问端所在名称空间;podSelector字段是用来描述被访问端的pod;如果这两个字段中只有一个字段前用了“-”,那么namespaceSelector和podSelector所指定的条件是与关系,即两个条件必须都满足;如果两个字段都用了“-”开头,则两个字段所指定的条件就是或关系,即满足其中一个字段的条件都可以被规则匹配;上述示例表示dev下的所有pod能够访问名称空间上的标签是name=def或name=dev的名称空间下的pod,并且这些pod的标签必须是app=nginx,rel=stable;即要想被dev名称空间下的pod所访问,对应pod的应该同时满足两个条件,第一对应名称空间上有name=def或者name=dev的标签;其次pod本身要有两个标签,app=nginx和rel=stable;这两个条件缺一不可;

   删除原有netpol,应用新的清单

[root@master01 ~]# kubectl get netpol -n dev NAME               POD-SELECTOR   AGE allow-dev-egress   <none>         28m [root@master01 ~]# kubectl delete netpol allow-dev-egress -n dev networkpolicy.networking.k8s.io "allow-dev-egress" deleted [root@master01 ~]# kubectl apply -f allow-dev-def-egress.yaml  networkpolicy.networking.k8s.io/allow-dev-egress created [root@master01 ~]# kubectl get netpol -n dev                     NAME               POD-SELECTOR   AGE allow-dev-egress   <none>         5s [root@master01 ~]# kubectl describe netpol allow-dev-egress -n dev Name:         allow-dev-egress Namespace:    dev Created on:   2021-01-04 19:44:32 +0800 CST Labels:       <none> Annotations:  <none> Spec:   PodSelector:     <none> (Allowing the specific traffic to all pods in this namespace)   Not affecting ingress traffic   Allowing egress traffic:     To Port: 80/TCP     To:       NamespaceSelector: name in (def,dev)       PodSelector: app=nginx,rel=stable   Policy Types: Egress [root@master01 ~]#  

  提示:可以看到对应的出站规则就有两个条件,首先称空间必须匹配,name=def或者name=dev,即对应名称空间必须要这样的标签;其次pod必须满足有app=nginx,rel=stable的标签;

  给default名称空间打上name=def的标签

[root@master01 ~]# kubectl get ns --show-labels NAME                   STATUS   AGE     LABELS default                Active   27d     <none> dev                    Active   4h24m   name=dev ingress-nginx          Active   13d     app.kubernetes.io/name=ingress-nginx,app.kubernetes.io/part-of=ingress-nginx kube-node-lease        Active   27d     <none> kube-public            Active   27d     <none> kube-system            Active   27d     <none> kubernetes-dashboard   Active   2d5h    <none> prod                   Active   57m     name=prod [root@master01 ~]# kubectl label ns default name=def namespace/default labeled [root@master01 ~]# kubectl get ns --show-labels      NAME                   STATUS   AGE     LABELS default                Active   27d     name=def dev                    Active   4h24m   name=dev ingress-nginx          Active   13d     app.kubernetes.io/name=ingress-nginx,app.kubernetes.io/part-of=ingress-nginx kube-node-lease        Active   27d     <none> kube-public            Active   27d     <none> kube-system            Active   27d     <none> kubernetes-dashboard   Active   2d5h    <none> prod                   Active   58m     name=prod [root@master01 ~]#  

  验证:用dev下的pod访问default和dev名称空间下pod标签为app=nginx,rel=stable的pod,看看对应pod是否能够被访问到?

[root@master01 ~]# kubectl get pods -n dev -o wide --show-labels NAME                        READY   STATUS    RESTARTS   AGE    IP            NODE             NOMINATED NODE   READINESS GATES   LABELS nginx-dep-cb74c595f-27qjp   1/1     Running   2          116m   10.244.1.18   node01.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-b92s4   1/1     Running   1          116m   10.244.4.17   node04.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-wdqnh   1/1     Running   1          116m   10.244.1.19   node01.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable [root@master01 ~]# kubectl get pods  -o wide --show-labels       NAME                        READY   STATUS    RESTARTS   AGE     IP            NODE             NOMINATED NODE   READINESS GATES   LABELS nginx-dep-cb74c595f-5mwl2   1/1     Running   1          117m    10.244.4.18   node04.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-vxc9z   1/1     Running   1          117m    10.244.3.12   node03.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-z265t   1/1     Running   1          117m    10.244.2.13   node02.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable web-0                       1/1     Running   8          3d21h   10.244.2.14   node02.k8s.org   <none>           <none>            app=nginx,controller-revision-hash=web-6db9455fdf,statefulset.kubernetes.io/pod-name=web-0 web-1                       1/1     Running   7          3d21h   10.244.4.16   node04.k8s.org   <none>           <none>            app=nginx,controller-revision-hash=web-6db9455fdf,statefulset.kubernetes.io/pod-name=web-1 web-2                       1/1     Running   8          3d21h   10.244.3.11   node03.k8s.org   <none>           <none>            app=nginx,controller-revision-hash=web-6db9455fdf,statefulset.kubernetes.io/pod-name=web-2 [root@master01 ~]# kubectl exec -it -n dev nginx-dep-cb74c595f-27qjp -- /bin/sh / # wget --spider --timeout=1 10.244.4.17 Connecting to 10.244.4.17 (10.244.4.17:80) / # wget --spider --timeout=1 10.244.1.19  Connecting to 10.244.1.19 (10.244.1.19:80) / # wget --spider --timeout=1 10.244.4.18 Connecting to 10.244.4.18 (10.244.4.18:80) / # wget --spider --timeout=1 10.244.3.12 Connecting to 10.244.3.12 (10.244.3.12:80) / # wget --spider --timeout=1 10.244.2.13 Connecting to 10.244.2.13 (10.244.2.13:80) / # wget --spider --timeout=1 10.244.2.14 Connecting to 10.244.2.14 (10.244.2.14:80) wget: download timed out / #  

  提示:可以看到用dev名称空间的pod访问名称空间上有name=def或name=dev标签并且对应pod上有app=nginx和rel=stable的标签的pod是能够正常访问到,如果不能满足对应名称空间和pod标签的pod就不能被访问;上面示例中最后一个podip不满足pod的标签,所以访问不到对应pod;

  示例:允许dev名称空间下的pod访问名称空间上有name=def或name=dev标签的名称空间下的pod或者名称空间上有name=prod标签,并且对应名称空间下的pod必须有app=nginx,rel=stable的标签;

[root@master01 ~]# cat allow-dev-def-egress.yaml apiVersion: networking.k8s.io/v1 kind: NetworkPolicy metadata:   name: allow-dev-egress   namespace: dev spec:   podSelector: {}    egress:     - to:       - namespaceSelector:           matchExpressions:           - key: name             operator: In             values: ["def","dev"]     - to:       - namespaceSelector:           matchLables:             name: prod         podSelector:           matchLabels:             app: nginx             rel: stable       ports:       - port: 80         protocol: TCP    policyTypes:   - Egress [root@master01 ~]#  

  应用配置清单

[root@master01 ~]# kubectl apply -f allow-dev-def-egress.yaml networkpolicy.networking.k8s.io/allow-dev-egress created [root@master01 ~]# kubectl get netpol -n dev NAME               POD-SELECTOR   AGE allow-dev-egress   <none>         10s [root@master01 ~]# kubectl describe netpol allow-dev-egress -n dev Name:         allow-dev-egress Namespace:    dev Created on:   2021-01-04 20:44:42 +0800 CST Labels:       <none> Annotations:  <none> Spec:   PodSelector:     <none> (Allowing the specific traffic to all pods in this namespace)   Not affecting ingress traffic   Allowing egress traffic:     To Port: <any> (traffic allowed to all ports)     To:       NamespaceSelector: name in (def,dev)     ----------     To Port: 80/TCP     To:       NamespaceSelector: name=prod       PodSelector: app=nginx,rel=stable   Policy Types: Egress [root@master01 ~]#  

  验证:用dev名称空间下的pod访问dev名称空间下的pod,看看是否可以正常访问?

[root@master01 ~]# kubectl get pods -n dev -o wide  NAME                        READY   STATUS    RESTARTS   AGE    IP            NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-27qjp   1/1     Running   2          168m   10.244.1.18   node01.k8s.org   <none>           <none> nginx-dep-cb74c595f-b92s4   1/1     Running   1          168m   10.244.4.17   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-wdqnh   1/1     Running   1          168m   10.244.1.19   node01.k8s.org   <none>           <none> [root@master01 ~]# kubectl exec -it -n dev nginx-dep-cb74c595f-27qjp -- /bin/sh / # wget --spider --timeout=1 10.244.4.17 Connecting to 10.244.4.17 (10.244.4.17:80) / # wget --spider --timeout=1 10.244.1.19 Connecting to 10.244.1.19 (10.244.1.19:80) / # exit [root@master01 ~]#  

  提示:可以看到dev名称空间下的pod相互访问没有问题;

  验证:用dev名称空间下的pod访问default名称空间下的pod ,看看是否可以正常访问?

[root@master01 ~]# kubectl get pods -n dev -o wide NAME                        READY   STATUS    RESTARTS   AGE    IP            NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-27qjp   1/1     Running   2          170m   10.244.1.18   node01.k8s.org   <none>           <none> nginx-dep-cb74c595f-b92s4   1/1     Running   1          170m   10.244.4.17   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-wdqnh   1/1     Running   1          170m   10.244.1.19   node01.k8s.org   <none>           <none> [root@master01 ~]# kubectl get pods -o wide NAME                        READY   STATUS    RESTARTS   AGE     IP            NODE             NOMINATED NODE   READINESS GATES nginx-dep-cb74c595f-5mwl2   1/1     Running   1          171m    10.244.4.18   node04.k8s.org   <none>           <none> nginx-dep-cb74c595f-vxc9z   1/1     Running   1          171m    10.244.3.12   node03.k8s.org   <none>           <none> nginx-dep-cb74c595f-z265t   1/1     Running   1          171m    10.244.2.13   node02.k8s.org   <none>           <none> web-0                       1/1     Running   8          3d22h   10.244.2.14   node02.k8s.org   <none>           <none> web-1                       1/1     Running   7          3d22h   10.244.4.16   node04.k8s.org   <none>           <none> web-2                       1/1     Running   8          3d22h   10.244.3.11   node03.k8s.org   <none>           <none> [root@master01 ~]# kubectl exec -it -n dev nginx-dep-cb74c595f-27qjp -- /bin/sh / # wget --spider --timeout=1 10.244.4.18 Connecting to 10.244.4.18 (10.244.4.18:80) / # wget --spider --timeout=1 10.244.3.12 Connecting to 10.244.3.12 (10.244.3.12:80) / # wget --spider --timeout=1 10.244.2.13 Connecting to 10.244.2.13 (10.244.2.13:80) / # wget --spider --timeout=1 10.244.2.14 Connecting to 10.244.2.14 (10.244.2.14:80) / # wget --spider --timeout=1 10.244.4.16 Connecting to 10.244.4.16 (10.244.4.16:80) wget: server returned error: HTTP/1.1 403 Forbidden / # wget --spider --timeout=1 10.244.3.11 Connecting to 10.244.3.11 (10.244.3.11:80) wget: server returned error: HTTP/1.1 403 Forbidden / #  

  提示:可以看到default名称空间下的所有pod都能够被dev名称空间下的pod所访问;后面显示403是因为对应web服务没有主页;

  验证:使用dev名称空间下的pod访问prod名称空间下的pod,看看对应pod是否能够被访问?

[root@master01 ~]# kubectl get ns prod --show-labels NAME   STATUS   AGE    LABELS prod   Active   118m   name=prod [root@master01 ~]# kubectl get pods -n prod -o wide --show-labels NAME                         READY   STATUS    RESTARTS   AGE    IP            NODE             NOMINATED NODE   READINESS GATES   LABELS nginx-dep-cb74c595f-rzbv2    1/1     Running   0          118m   10.244.4.19   node04.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-v8ssx    1/1     Running   0          118m   10.244.3.13   node03.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep-cb74c595f-zsqcc    1/1     Running   0          118m   10.244.2.15   node02.k8s.org   <none>           <none>            app=nginx,pod-template-hash=cb74c595f,rel=stable nginx-dep1-f85fdcdbc-kdq5b   1/1     Running   0          117m   10.244.1.21   node01.k8s.org   <none>           <none>            app=nginx,pod-template-hash=f85fdcdbc nginx-dep1-f85fdcdbc-n8cvs   1/1     Running   0          117m   10.244.3.14   node03.k8s.org   <none>           <none>            app=nginx,pod-template-hash=f85fdcdbc nginx-dep1-f85fdcdbc-vz2mp   1/1     Running   0          117m   10.244.4.20   node04.k8s.org   <none>           <none>            app=nginx,pod-template-hash=f85fdcdbc [root@master01 ~]# kubectl get pods -n dev NAME                        READY   STATUS    RESTARTS   AGE nginx-dep-cb74c595f-27qjp   1/1     Running   2          174m nginx-dep-cb74c595f-b92s4   1/1     Running   1          174m nginx-dep-cb74c595f-wdqnh   1/1     Running   1          174m [root@master01 ~]# kubectl exec -it -n dev nginx-dep-cb74c595f-27qjp -- /bin/sh / # wget --spider --timeout=1 10.244.4.20  Connecting to 10.244.4.20 (10.244.4.20:80) wget: download timed out / # wget --spider --timeout=1 10.244.3.14  Connecting to 10.244.3.14 (10.244.3.14:80) wget: download timed out / # wget --spider --timeout=1 10.244.1.21 Connecting to 10.244.1.21 (10.244.1.21:80) wget: download timed out / # wget --spider --timeout=1 10.244.2.15 Connecting to 10.244.2.15 (10.244.2.15:80) / # wget --spider --timeout=1 10.244.3.13 Connecting to 10.244.3.13 (10.244.3.13:80) / # wget --spider --timeout=1 10.244.4.19 Connecting to 10.244.4.19 (10.244.4.19:80) / #  

  提示:可以看到在prod名称空间下对应pod标签为app=nginx,rel=stable的pod都能够被dev下的pod所访问,如果对应标签不是app=nginx,rel=stable的pod则不能被访问到;

  从上面的示例可以总结,在k8s上网络策略是白名单机制,所谓白名单机制是指,只有明确定义的策略才会被允许放行,默认没有指定的规则就是拒绝的,即条件不匹配的都会被拒绝;其次对于ingress或egress来说,对应的from或to都是用来指定访问端或被访问端的信息;如果我们在对应的字段中没有定义namespaceSelector字段,默认ingress或egrss会匹配当前netpol所在名称空间,即在没有明确指定namespaceSelector字段时,对应的其他条件都是针对当前netpol所在名称空间;多个条件组合使用,如果多个条件都在一个列表中,则表示多个条件间是与关系,即指定的条件需要同时满足对应策略才会放行;如果多个条件不再同一个列表中,则多个条件之间是或关系,即满足其中一个条件都会被对应策略放行;

赞(0) 打赏
未经允许不得转载:张拓的天空 » 容器编排系统K8s之NetworkPolicy资源
分享到: 更多 (0)

评论 抢沙发

  • 昵称 (必填)
  • 邮箱 (必填)
  • 网址

专业的IT技术经验分享 更专业 更方便

联系我们本站主机

觉得文章有用就打赏一下文章作者

支付宝扫一扫打赏

微信扫一扫打赏