Afaik a K8s NetworkPolicy
can only allow pods matching a label to do something. I do not want to
- Deny all traffic
- Allow traffic for all pods except the ones matching my label
, but instead
- Allow all traffic
- Deny traffic for pods matching my label
How do I do that?
From kubectl explain NetworkPolicy.spec.ingress.from
DESCRIPTION:
List of sources which should be able to access the pods selected for this
rule. Items in this list are combined using a logical OR operation. If this
field is empty or missing, this rule matches all sources (traffic not
restricted by source). If this field is present and contains at least one
item, this rule allows traffic only if the traffic matches at least one
item in the from list.
As far as I understand this, we can only allow, not deny.
CodePudding user response:
As you mentioned in the comments, you are using the Kind tool for running Kubernetes. Instead of kindnet CNI plugin (default CNI plugin for Kind) which does not support Kubernetes network policies, you can use Calico CNI plugin which support Kubernetes network policies it has its own, similar solution called Calico network policies.
Example - I will create cluster with disabled default kind CNI plugin enabled NodePort for testing (assuming that you have kind
kubectl
tools already installed):
kind-cluster-config.yaml file:
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
networking:
disableDefaultCNI: true # disable kindnet
podSubnet: 192.168.0.0/16 # set to Calico's default subnet
nodes:
- role: control-plane
extraPortMappings:
- containerPort: 30000
hostPort: 30000
listenAddress: "0.0.0.0" # Optional, defaults to "0.0.0.0"
protocol: tcp # Optional, defaults to tcp
Time for create a cluster using above config:
kind create cluster --config kind-cluster-config.yaml
When cluster is ready, I will install Calico CNI plugin:
kubectl apply -f https://docs.projectcalico.org/manifests/calico.yaml
I will wait until all calico pods are ready (kubectl get pods -n kube-system
command to check). Then, I will create sample nginx deployment service type NodePort for accessing:
nginx-deploy-service.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx-deployment
spec:
selector:
matchLabels:
app: nginx
replicas: 2 # tells deployment to run 2 pods matching the template
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.14.2
ports:
- containerPort: 80
---
apiVersion: v1
kind: Service
metadata:
name: nginx-service
spec:
type: NodePort
selector:
app: nginx
ports:
- protocol: TCP
port: 80
targetPort: 80
nodePort: 30000
Let's apply it: kubectl apply -f nginx-deploy-service.yaml
So far so good. Now I will try to access nginx-service
using node IP (kubectl get nodes -o wide
command to check node IP address):
curl 172.18.0.2:30000
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
...
Okay, it's working.
Now time to install calicoctl
and apply some example policy - based on this tutorial - to block ingress traffic only for pods with label app
with value nginx
:
calico-rule.yaml:
apiVersion: projectcalico.org/v3
kind: GlobalNetworkPolicy
metadata:
name: default-deny
spec:
selector: app == "nginx"
types:
- Ingress
Apply it:
calicoctl apply -f calico-rule.yaml
Successfully applied 1 'GlobalNetworkPolicy' resource(s)
Now I can't reach the address 172.18.0.2:30000
which was working previously. The policy is working fine!
Read more about calico policies:
Also check this GitHub topic for more information about NetworkPolicy support in Kind.
EDIT:
Seems like Calico plugin supports as well Kubernetes NetworkPolicy, so you can just install Calico CNI plugin and apply the following policy:
kind: NetworkPolicy
apiVersion: networking.k8s.io/v1
metadata:
name: default-deny
spec:
podSelector:
matchLabels:
app: nginx
policyTypes:
- Ingress
I tested it and seems it's working fine as well.