Step 1: Expose Apigee ingress gateway

This step explains how to set up the Kubernetes service for your Apigee ingress gateway. The Kubernetes service is needed for each ingress deployment to obtain an IP address that can be exposed. Client calls to proxies will invoke a hostname that will resolve to this IP address.

Options for Kubernetes services for your Apigee ingress gateway

There are two options for providing a kubernetes service to assign the IP address ...

Create a custom Kubernetes service for each Apigee ingress gateway (recommended).
Use the default Kubernetes service.

Create your own Kubernetes service

For production environments, Apigee recommends you create a custom Kubernetes service for each ingress gateway.

Apigee deletes default service, but does not delete the custom kubernetes service upon cleanup. Hence, the IP address won't be released upon reinstallation of Apigee hybrid.
For platforms not on Google Cloud, like EKS, AKS, and OpenShift, you need to customize the Kubernetes service to work with the cloud provider. Therefore it is better to create a custom Kubernetes service than to use the default service, because apigee does not support all customizations to the default Kubernetes service.

Use the following steps to set up and route traffic to the new ingress gateway.

Create a Kubernetes service with the required pod selector labels, app, ingress_name, and org. These labels are already present in Apigee ingress gateway pods.

Create a service file using the following template:

apiVersion: v1
kind: Service
metadata:
  name: SERVICE_NAME
  namespace: APIGEE_NAMESPACE
spec:
  ports:
  - name: status-port
    port: 15021
    protocol: TCP
    targetPort: 15021
  - name: https
    port: 443
    protocol: TCP
    targetPort: 8443
  selector:
    app: apigee-ingressgateway #required
    ingress_name: INGRESS_NAME
    org: ORG_NAME
  type: LoadBalancer
  loadBalancerIP: LOAD_BALANCER_IP

SERVICE_NAME is a name used to identify this service. For example, apigee-prod-1.
Note: To more easily identify your services, use the ingress name as part of the service name.
INGRESS_NAME is the name of this Apigee ingress gateway. It must match the name you provided for ingressGateways.name in your overrides.yaml file. For more information see ingressGateways in the Configuration properties reference.
ORG_NAME is the name of the Apigee organization. It must match the name you provided for org in your overrides.yaml file. For more information see org in the Configuration properties reference.
LOAD_BALANCER_IP is the IP address for the load balancer.
If you have an IP address reserved for your load balancer, enter that.

If you do not have an IP address already reserved, you can use the one returned by the default Apigee ingress service. Use the EXTERNAL-IP returned by the following command:
```
kubectl get svc -n APIGEE_NAMESPACE -l app=apigee-ingressgateway
```
Caution: Because of a limitation on Google Cloud internal Application Load Balancer, this Apigee routing option does not support HTTP 1.0 requests. Incoming client requests that specify the HTTP 1.0 protocol will fail. Later HTTP versions are supported.

Apigee ingress gateway exposes the following ports:

Port	Description
443	Runtime traffic.
15021	Health check. `status-port` exposes a `/healthz/ready` endpoint that can be used with GKE Ingress health checks.

Create the service by applying the SERVICE_FILENAME.yaml:
```
kubectl apply -f SERVICE_FILENAME.yaml
```

Find the external IP of Apigee ingress gateway with the following command:

kubectl get svc -n APIGEE_NAMESPACE SERVICE_NAME

Your output should look something like:

NAME                                        TYPE           CLUSTER-IP    EXTERNAL-IP     PORT(S)                                      AGE
apigee-ingressgateway-prod-hybrid-37a39bd   LoadBalancer   192.0.2.123   233.252.0.123   15021:32049/TCP,80:31624/TCP,443:30723/TCP   16h

Disable the loadbalancer for the default Apigee ingress gateway service:
1. Update the ingressGateways[].svcType property to ClusterIP in your overrides file:
```
ingressGateways:
  svcType: ClusterIP
```
2. Upgrade the apigee-org chart to apply the changes:
```
helm upgrade ORG_NAME apigee-org/ \
  --install \
  --namespace apigee \
  --atomic \
  -f overrides.yaml
```
See Disable the loadbalancer for the default Apigee ingress gateway service for more information.
Test the ingress gateway by making a healthcheck call.
```
curl -H 'User-Agent: GoogleHC' https://DOMAIN/healthz/ingress -k \
  --resolve "DOMAIN:443:INGRESS_IP_ADDRESS"
```
Where
- DOMAIN is the domain you provided as the hostname for the environment group you created in Project and org setup - Step 3: Create an environment group.
- INGRESS_IP_ADDRESS is the ingress IP address, for example 233.252.0.123.
On success the command returns:
```
Apigee Ingress is healthy
```
Use this IP address to update your DNS record (usually an A or CNAME record) at your registrar or DNS provider.

Use the default Kubernetes service

For non-production environments or to test initial traffic through the Apigee ingress gateway, Apigee hybrid provides default Kubernetes service for each ingress deployment.

You can make limited configuration changes to the default service in your overrides.yaml file. For the available configuration options, see Managing Apigee ingress gateway. For example, you can add annotations.

For production environments, it is recommended you provide a Kubernetes service for ingress. Follow the steps in Create your own Kubernetes service.

Find the external IP of the default Apigee ingress service with the following command:

kubectl get svc -n APIGEE_NAMESPACE -l app=apigee-ingressgateway