Skip to content

howto_load_balancer.md

Latest commit

 

History

History
119 lines (92 loc) · 4.06 KB

howto_load_balancer.md

File metadata and controls

119 lines (92 loc) · 4.06 KB

Deploying a TCP Load Balancer for User Workloads

MicroShift does not currently offer an implementation of network load balancers (Services of type LoadBalancer). If load balancer facilities are required by user workloads, it is possible to deploy 3rd party load balancer services.

This document demonstates how to deploy the MetalLB service, which is a load balancer implementation for bare metal clusters.

Create MicroShift Server

Use the instructions in the Install MicroShift on RHEL for Edge document to configure a virtual machine running MicroShift.

Log into the virtual machine and run the following commands to configure the MicroShift access and check if the PODs are up and running.

mkdir ~/.kube
sudo cat /var/lib/microshift/resources/kubeadmin/kubeconfig > ~/.kube/config
oc get pods -A

Install Load Balancer

Log into the virtual machine and run the following commands to create the MetalLB namespace and deployment.

oc apply -f https://raw.githubusercontent.com/metallb/metallb/v0.13.4/config/manifests/metallb-native.yaml

Verify that the MetalLB pods are up and running in the metallb-system namespace.

oc get pods -n metallb-system
NAME                          READY   STATUS    RESTARTS   AGE
controller-64cc46b9f9-2csb7   1/1     Running   0          107s
speaker-fqmq4                 1/1     Running   0          107s

Once all the pods are available, create an IPAddressPool resource to define the default address pool for the load balancer to use.

oc create -f - <<EOF
apiVersion: metallb.io/v1beta1
kind: IPAddressPool
metadata:
  name: default
  namespace: metallb-system
spec:
  addresses:
  - 192.168.1.240-192.168.1.250
EOF

Install User Workload

Log into the virtual machine and create the namespace to be used for deploying the test application.

NAMESPACE=nginx-lb-test
oc create ns $NAMESPACE

Run the following command to deploy 3 replicas of a test nginx application in the specified namespace.

oc apply -n $NAMESPACE -f https://raw.githubusercontent.com/openshift/microshift/main/docs/config/nginx-IP-header.yaml

The application is configured to return the X-Server-IP header with the container IP address.

Verify that all the 3 replicas of the application started successfully.

oc get pods -n $NAMESPACE

Create Load Balancer Service

Log into the virtual machine and run the following command to create the LoadBalancer service for nginx application using the default MetalLB address pool.

oc create -n $NAMESPACE -f - <<EOF
apiVersion: v1
kind: Service
metadata:
  name: nginx
  annotations:
    metallb.universe.tf/address-pool: default
spec:
  ports:
  - port: 80
    targetPort: 80
  selector:
    app: nginx
  type: LoadBalancer
EOF

Verify that the service exists and an external IP address has been assigned to it.

oc get svc -n $NAMESPACE
NAME    TYPE           CLUSTER-IP      EXTERNAL-IP     PORT(S)        AGE
nginx   LoadBalancer   10.43.183.104   192.168.1.241   80:32434/TCP   2m

Test Load Balancer

Log into the virtual machine and run the following commands to verify that the load balancer distributes requests among all the running application instances.

Set the EXTERNAL_IP environment variable to the external IP of the LoadBalancer service.

EXTERNAL_IP=192.168.1.241
seq 5 | xargs -Iz curl -s -I http://$EXTERNAL_IP | grep X-Server-IP

The above command attempts to perform five connections to the nginx application using the external IP address of the LoadBalancer service. Only the value of the X-Server-IP header is filtered from the application response.

The output of this command should contain different IP addresses, demonstrating that the load balancer service works by distributing the traffic among the instances of the application.

X-Server-IP: 10.42.0.41
X-Server-IP: 10.42.0.41
X-Server-IP: 10.42.0.43
X-Server-IP: 10.42.0.41
X-Server-IP: 10.42.0.43