EKS/loxilb external mode
Create an EKS cluster with ingress access enabled by loxilb (external-mode)
This document details the steps to create an EKS cluster and allow external ingress access using loxilb running in external mode. loxilb will run as EC2 instances in EKS cluster's VPC while loxilb's operator, kube-loxilb, will run as a replica-set inside EKS cluster.
Create EKS cluster with 4 worker nodes from a bastion node inside your VPC
- It is assumed that aws-cli, kubectl and eksctl are installed in a bastion node
$ eksctl create cluster --version 1.24 --name loxilb-demo --vpc-nat-mode Single --region ap-northeast-2 --node-type t3.small --nodes 4 --with-oidc --managed
- Create kube config for kubectl access
$ aws eks update-kubeconfig --region ap-northeast-2 --name loxilb-demo - Double confirm the cluster created
$ kubectl get pods -A NAMESPACE NAME READY STATUS RESTARTS AGE kube-system aws-node-2fpm4 2/2 Running 0 14m kube-system aws-node-6vhlr 2/2 Running 0 14m kube-system aws-node-9kzb2 2/2 Running 0 14m kube-system aws-node-vvkq5 2/2 Running 0 14m kube-system coredns-5ff5b8d45c-gj9kj 1/1 Running 0 21m kube-system coredns-5ff5b8d45c-p64fd 1/1 Running 0 21m kube-system kube-proxy-5j9gf 1/1 Running 0 14m kube-system kube-proxy-5tm8w 1/1 Running 0 14m kube-system kube-proxy-894k9 1/1 Running 0 14m kube-system kube-proxy-xgfb8 1/1 Running 0 14m
Deploy loxilb as EC2 instances in EKS's VPC
- Create a file
launch-loxilb.shwith the following contents (in bastion node)#!/bin/bash sudo apt-get update && apt-get install -y snapd sudo snap install docker sleep 30 sudo docker run -u root --cap-add SYS_ADMIN --net=host --restart unless-stopped --privileged -dit -v /dev/log:/dev/log --name loxilb ghcr.io/loxilb-io/loxilb:latest - Deploy loxilb ec2 instance(s) using the above init-script
$ aws ec2 run-instances --image-id ami-01ed8ade75d4eee2f --count 1 --instance-type t3.medium --key-name aws-netlox --security-group-ids sg-0e2638db05b256476 --subnet-id subnet-0109b973f5f674f99 --associate-public-ip-address --user-data file://launch-loxilb.sh
Note : subnet-id should be any subnet with public access enabled from the EKS cluster. Rest of the args can be changed as applicable
- Double confirm loxilb EC2 instances are running properly in amazon aws console or using aws cli.
- Disable source/dest check of the loxilb EC2 instances
aws ec2 modify-network-interface-attribute --network-interface-id eni-02e1cbfa022eb0901 --no-source-dest-check
Deploy loxilb's operator (kube-loxilb)
- Create a file kube-loxilb.yml with the following contents
--- --- apiVersion: v1 kind: ServiceAccount metadata: name: kube-loxilb namespace: kube-system --- kind: ClusterRole apiVersion: rbac.authorization.k8s.io/v1 metadata: name: kube-loxilb rules: - apiGroups: - "" resources: - nodes verbs: - get - watch - list - patch - apiGroups: - "" resources: - pods verbs: - get - watch - list - patch - apiGroups: - "" resources: - endpoints - services - namespaces - services/status verbs: - get - watch - list - patch - update - apiGroups: - gateway.networking.k8s.io resources: - gatewayclasses - gatewayclasses/status - gateways - gateways/status - tcproutes - udproutes verbs: ["get", "watch", "list", "patch", "update"] - apiGroups: - discovery.k8s.io resources: - endpointslices verbs: - get - watch - list - apiGroups: - apiextensions.k8s.io resources: - customresourcedefinitions verbs: - get - watch - list - apiGroups: - authentication.k8s.io resources: - tokenreviews verbs: - create - apiGroups: - authorization.k8s.io resources: - subjectaccessreviews verbs: - create - apiGroups: - bgppeer.loxilb.io resources: - bgppeerservices verbs: - get - watch - list - create - update - delete - apiGroups: - bgppolicydefinedsets.loxilb.io resources: - bgppolicydefinedsetsservices verbs: - get - watch - list - create - update - delete - apiGroups: - bgppolicydefinition.loxilb.io resources: - bgppolicydefinitionservices verbs: - get - watch - list - create - update - delete - apiGroups: - bgppolicyapply.loxilb.io resources: - bgppolicyapplyservices verbs: - get - watch - list - create - update - delete - apiGroups: - loxiurl.loxilb.io resources: - loxiurls verbs: - get - watch - list - create - update - delete --- kind: ClusterRoleBinding apiVersion: rbac.authorization.k8s.io/v1 metadata: name: kube-loxilb roleRef: apiGroup: rbac.authorization.k8s.io kind: ClusterRole name: kube-loxilb subjects: - kind: ServiceAccount name: kube-loxilb namespace: kube-system --- apiVersion: apps/v1 kind: Deployment metadata: name: kube-loxilb namespace: kube-system labels: app: kube-loxilb-app spec: replicas: 1 selector: matchLabels: app: kube-loxilb-app template: metadata: labels: app: kube-loxilb-app spec: hostNetwork: true dnsPolicy: ClusterFirstWithHostNet tolerations: # Mark the pod as a critical add-on for rescheduling. - key: CriticalAddonsOnly operator: Exists priorityClassName: system-node-critical serviceAccountName: kube-loxilb terminationGracePeriodSeconds: 0 containers: - name: kube-loxilb image: ghcr.io/loxilb-io/kube-loxilb:latest imagePullPolicy: Always command: - /bin/kube-loxilb args: - --loxiURL=http://192.168.31.175:11111 - --externalCIDR=0.0.0.0/32 - --setLBMode=5 #- --setRoles:0.0.0.0 resources: requests: cpu: "100m" memory: "50Mi" limits: cpu: "100m" memory: "50Mi" securityContext: privileged: true capabilities: add: ["NET_ADMIN", "NET_RAW"]
Note1: --externalCIDR args can be set to any Public IP address via which any of the worker nodes can be accessed. It can be also set to simply 0.0.0.0/32 which means LB will be performed on any of the nodes where loxilb runs. The decision of which loxilb node/instance will be chosen as ingress in this case can be done by Route53/DNS.
Note2: --loxiURL args should be set to privateIP address(es) of the loxilb ec2 instances accessible from the EKS cluster. Currently, kube-loxilb can't autodetect the EC2 instances running loxilb in external mode.
- Deploy kube-loxilb to EKS cluster
$ kubectl apply -f kube-loxilb.yml serviceaccount/kube-loxilb created clusterrole.rbac.authorization.k8s.io/kube-loxilb created deployment.apps/kube-loxilb created - Check the state of the EKS cluster
$ kubectl get pods -A NAMESPACE NAME READY STATUS RESTARTS AGE kube-system aws-node-2fpm4 2/2 Running 0 14m kube-system aws-node-6vhlr 2/2 Running 0 14m kube-system aws-node-9kzb2 2/2 Running 0 14m kube-system aws-node-vvkq5 2/2 Running 0 14m kube-system coredns-5ff5b8d45c-gj9kj 1/1 Running 0 21m kube-system coredns-5ff5b8d45c-p64fd 1/1 Running 0 21m kube-system kube-proxy-5j9gf 1/1 Running 0 14m kube-system kube-proxy-5tm8w 1/1 Running 0 14m kube-system kube-proxy-894k9 1/1 Running 0 14m kube-system kube-proxy-xgfb8 1/1 Running 0 14m kube-system kube-loxilb-6477d6897f-vz74f 1/1 Running 0 5m
Install a test service
- Create a file nginx.yml with the following contents:
apiVersion: v1 kind: Service metadata: name: nginx-lb1 annotations: loxilb.io/usepodnetwork : "yes" spec: externalTrafficPolicy: Local loadBalancerClass: loxilb.io/loxilb selector: what: nginx-test ports: - port: 55002 targetPort: 80 type: LoadBalancer --- apiVersion: v1 kind: Pod metadata: name: nginx-test labels: what: nginx-test spec: containers: - name: nginx-test image: nginx:stable ports: - containerPort: 80 - Please not the usage of annotation
loxilb.io/usepodnetwork : "yes". This would imply loxilb will directly use PodIP and TargetPort to reach out as its end-points. This feature is only available with EKS currently and should provide additional performance boost. - Deploy test nginx service to EKS
$ kubectl apply -f nginx.yml service/nginx-lb1 created -
Check the state of the EKS cluster
$ kubectl get pods -A NAMESPACE NAME READY STATUS RESTARTS AGE default nginx-test 1/1 Running 0 50s kube-system aws-node-2fpm4 2/2 Running 0 14m kube-system aws-node-6vhlr 2/2 Running 0 14m kube-system aws-node-9kzb2 2/2 Running 0 14m kube-system aws-node-vvkq5 2/2 Running 0 14m kube-system coredns-5ff5b8d45c-gj9kj 1/1 Running 0 21m kube-system coredns-5ff5b8d45c-p64fd 1/1 Running 0 21m kube-system kube-proxy-5j9gf 1/1 Running 0 14m kube-system kube-proxy-5tm8w 1/1 Running 0 14m kube-system kube-proxy-894k9 1/1 Running 0 14m kube-system kube-proxy-xgfb8 1/1 Running 0 14m kube-system kube-loxilb-6477d6897f-vz74f 1/1 Running 0 5m -
Check the external service for service ingress (via loxilb)
$ kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.100.0.1 <none> 443/TCP 10h nginx-lb1 LoadBalancer 10.100.244.105 llbanyextip 55005:30055/TCP 24s
Test the service
- Try to access the service from outside (internet). We can use any public IP associated with any of the loxilb ec2 instances
$ curl http://3.37.191.xx:55005
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
<style>
html { color-scheme: light dark; }
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>
Note - We would need to make sure AWS security groups are setup properly to allow access for ingress traffic.
Restricting loxilb service for a local-zone node-group
For limiting loxilb services to a specific node group of a local-zone, we can use kubenetes node-labels to limit the endpoints of that service to that node-group only. For example, if all the nodes in a local-zone node-groups have a label node.kubernetes.io/local-zone2=true, then we can create a loxilb service with a following annotation :
apiVersion: v1
kind: Service
metadata:
name: nginx-lb1
annotations:
loxilb.io/nodelabel: "node.kubernetes.io/local-zone2"
spec:
externalTrafficPolicy: Local
loadBalancerClass: loxilb.io/loxilb
selector:
what: nginx-test
ports:
- port: 55002
targetPort: 80
type: LoadBalancer