* Docs: update sidebar * Docs: move registry documentation into docs/ * Docs: move rbd_provisioner documentation into docs/ * Docs: move cephfs_provisioner into docs/ * Docs: move local_volume_provisioner documentation into docs/ * Docs: move ambassador.md to docs/ingress_controller/ * Docs: move metallb.md to docs/ingress_controller/ * Docs: move ingress_nginx documentation into docs/ * Docs: move alb_ingress_controller documentation into docs/ * Docs: merge ambassador documentation into docs/ingress_controller/ * Docs: move cert_manager documentation into docs/ * Docs: move bootstrap-os documentation into docs/ * Docs: update file locations in sidebar
7.8 KiB
Installation Guide
Contents
Prerequisite Generic Deployment Command
!!! attention
The default configuration watches Ingress object from all the namespaces.
To change this behavior use the flag --watch-namespace
to limit the scope to a particular namespace.
!!! warning If multiple Ingresses define different paths for the same host, the ingress controller will merge the definitions.
!!! attention If you're using GKE you need to initialize your user as a cluster-admin with the following command:
kubectl create clusterrolebinding cluster-admin-binding \
--clusterrole cluster-admin \
--user $(gcloud config get-value account)
The following Mandatory Command is required for all deployments except for AWS. See below for the AWS version.
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v0.40.2/deploy/static/provider/cloud/deploy.yaml
Provider Specific Steps
There are cloud provider specific yaml files.
Docker for Mac
Kubernetes is available in Docker for Mac (from version 18.06.0-ce)
First you need to enable kubernetes.
Then you have to create a service:
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/cloud-generic.yaml
minikube
For standard usage:
minikube addons enable ingress
For development:
- Disable the ingress addon:
minikube addons disable ingress
- Execute
make dev-env
- Confirm the
nginx-ingress-controller
deployment exists:
$ kubectl get pods -n ingress-nginx
NAME READY STATUS RESTARTS AGE
default-http-backend-66b447d9cf-rrlf9 1/1 Running 0 12s
nginx-ingress-controller-fdcdcd6dd-vvpgs 1/1 Running 0 11s
AWS
In AWS we use an Elastic Load Balancer (ELB) to expose the NGINX Ingress controller behind a Service of Type=LoadBalancer
.
Since Kubernetes v1.9.0 it is possible to use a classic load balancer (ELB) or network load balancer (NLB)
Please check the elastic load balancing AWS details page
Elastic Load Balancer - ELB
This setup requires to choose in which layer (L4 or L7) we want to configure the Load Balancer:
- Layer 4: Use an Network Load Balancer (NLB) with TCP as the listener protocol for ports 80 and 443.
- Layer 7: Use an Elastic Load Balancer (ELB) with HTTP as the listener protocol for port 80 and terminate TLS in the ELB
For L4:
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/deploy.yaml
For L7:
Change the value of service.beta.kubernetes.io/aws-load-balancer-ssl-cert
in the file provider/aws/deploy-tls-termination.yaml
replacing the dummy id with a valid one. The dummy value is "arn:aws:acm:us-west-2:XXXXXXXX:certificate/XXXXXX-XXXXXXX-XXXXXXX-XXXXXXXX"
Check that no change is necessary with regards to the ELB idle timeout. In some scenarios, users may want to modify the ELB idle timeout, so please check the ELB Idle Timeouts section for additional information. If a change is required, users will need to update the value of service.beta.kubernetes.io/aws-load-balancer-connection-idle-timeout
in provider/aws/deploy-tls-termination.yaml
Then execute:
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/deploy-tls-termination.yaml
This example creates an ELB with just two listeners, one in port 80 and another in port 443
ELB Idle Timeouts
In some scenarios users will need to modify the value of the ELB idle timeout.
Users need to ensure the idle timeout is less than the keepalive_timeout that is configured for NGINX.
By default NGINX keepalive_timeout
is set to 75s
.
The default ELB idle timeout will work for most scenarios, unless the NGINX keepalive_timeout has been modified,
in which case service.beta.kubernetes.io/aws-load-balancer-connection-idle-timeout
will need to be modified to ensure it is less than the keepalive_timeout
the user has configured.
Please Note: An idle timeout of 3600s
is recommended when using WebSockets.
More information with regards to idle timeouts for your Load Balancer can be found in the official AWS documentation.
Network Load Balancer (NLB)
This type of load balancer is supported since v1.10.0 as an ALPHA feature.
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/service-nlb.yaml
GCE-GKE
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/cloud-generic.yaml
Important Note: proxy protocol is not supported in GCE/GKE
Azure
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/cloud-generic.yaml
Bare-metal
Using NodePort:
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/baremetal/service-nodeport.yaml
!!! tip For extended notes regarding deployments on bare-metal, see Bare-metal considerations.
Verify installation
To check if the ingress controller pods have started, run the following command:
kubectl get pods --all-namespaces -l app.kubernetes.io/name=ingress-nginx --watch
Once the operator pods are running, you can cancel the above command by typing Ctrl+C
.
Now, you are ready to create your first ingress.
Detect installed version
To detect which version of the ingress controller is running, exec into the pod and run nginx-ingress-controller version
command.
POD_NAMESPACE=ingress-nginx
POD_NAME=$(kubectl get pods -n $POD_NAMESPACE -l app.kubernetes.io/component=controller -o jsonpath='{.items[0].metadata.name}')
kubectl exec -it $POD_NAME -n $POD_NAMESPACE -- /nginx-ingress-controller --version
Using Helm
NGINX Ingress controller can be installed via Helm using the chart ingress-nginx/ingress-nginx. Official documentation is here
To install the chart with the release name my-nginx
:
helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
helm install my-nginx ingress-nginx/ingress-nginx
Detect installed version:
POD_NAME=$(kubectl get pods -l app.kubernetes.io/name=ingress-nginx -o jsonpath='{.items[0].metadata.name}')
kubectl exec -it $POD_NAME -- /nginx-ingress-controller --version