Skip to content
Snippets Groups Projects
Commit beb47e1c authored by zhanwang's avatar zhanwang Committed by Kubernetes Prow Robot
Browse files

update ingress_nginx install guide (#5502)

parent 303c3654
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
roles/kubernetes-apps/ingress_controller/ingress_nginx/README.md
+ 118
− 191
View file @ beb47e1c
Installation Guide # Installation Guide
==================
## Contents
Contents
-------- - [Prerequisite Generic Deployment Command](#prerequisite-generic-deployment-command)
- [Provider Specific Steps](#provider-specific-steps)
- [Mandatory commands](#mandatory-commands) - [Docker for Mac](#docker-for-mac)
- [Install without RBAC roles](#install-without-rbac-roles) - [minikube](#minikube)
- [Install with RBAC roles](#install-with-rbac-roles) - [AWS](#aws)
- [Custom Provider](#custom-provider) - [GCE - GKE](#gce-gke)
- [minikube](#minikube) - [Azure](#azure)
- [AWS](#aws) - [Bare-metal](#bare-metal)
- [GCE - GKE](#gce---gke) - [Verify installation](#verify-installation)
- [Azure](#azure) - [Detect installed version](#detect-installed-version)
- [Baremetal](#baremetal) - [Using Helm](#using-helm)
- [Using Helm](#using-helm)
- [Verify installation](#verify-installation) ## Prerequisite Generic Deployment Command
- [Detect installed version](#detect-installed-version)
- [Deploying the config-map](#deploying-the-config-map) !!! attention
The default configuration watches Ingress object from *all the namespaces*.
Generic Deployment To change this behavior use the flag `--watch-namespace` to limit the scope to a particular namespace.
------------------
!!! warning
The following resources are required for a generic deployment. If multiple Ingresses define different paths for the same host, the ingress controller will merge the definitions.
### Mandatory commands !!! attention
If you're using GKE you need to initialize your user as a cluster-admin with the following command:
``` console ```console
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/namespace.yaml \ kubectl create clusterrolebinding cluster-admin-binding \
| kubectl apply -f - --clusterrole cluster-admin \
--user $(gcloud config get-value account)
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/default-backend.yaml \ ```
| kubectl apply -f -
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/configmap.yaml \
| kubectl apply -f -
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/tcp-services-configmap.yaml \ The following **Mandatory Command** is required for all deployments.
| kubectl apply -f -
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/udp-services-configmap.yaml \ ```console
| kubectl apply -f - kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/mandatory.yaml
``` ```
### Install without RBAC roles !!! tip
If you are using a Kubernetes version previous to 1.14, you need to change `kubernetes.io/os` to `beta.kubernetes.io/os` at line 217 of [mandatory.yaml](https://github.com/kubernetes/ingress-nginx/blob/master/deploy/static/mandatory.yaml#L217), see [Labels details](https://kubernetes.io/docs/reference/kubernetes-api/labels-annotations-taints/).
``` console ### Provider Specific Steps
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/without-rbac.yaml \
| kubectl apply -f -
```
### Install with RBAC roles There are cloud provider specific yaml files.
Please check the [RBAC](rbac.md) document. #### Docker for Mac
``` console Kubernetes is available in Docker for Mac (from [version 18.06.0-ce](https://docs.docker.com/docker-for-mac/release-notes/#stable-releases-of-2018))
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/rbac.yaml \
| kubectl apply -f -
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/with-rbac.yaml \ [enable]: https://docs.docker.com/docker-for-mac/#kubernetes
| kubectl apply -f -
```
Custom Service Provider Deployment Create a service
----------------------------------
There are cloud provider specific yaml files. ```console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/cloud-generic.yaml
```
### minikube #### minikube
For standard usage: For standard usage:
``` console ```console
minikube addons enable ingress minikube addons enable ingress
``` ```
For development: For development:
1. Disable the ingress addon: 1. Disable the ingress addon:
``` console
$ minikube addons disable ingress
```
2. Use the [docker daemon](https://github.com/kubernetes/minikube/blob/master/docs/reusing_the_docker_daemon.md)
3. [Build the image](../docs/development.md)
4. Perform [Mandatory commands](#mandatory-commands)
5. Install the `nginx-ingress-controller` deployment [without RBAC roles](#install-without-rbac-roles) or [with RBAC roles](#install-with-rbac-roles)
6. Edit the `nginx-ingress-controller` deployment to use your custom image. Local images can be seen by performing `docker images`.
``` console ```console
$ kubectl edit deployment nginx-ingress-controller -n ingress-nginx minikube addons disable ingress
``` ```
edit the following section:
``` yaml
image: <IMAGE-NAME>:<TAG>
imagePullPolicy: IfNotPresent
name: nginx-ingress-controller
```
7. Confirm the `nginx-ingress-controller` deployment exists: 2. Execute `make dev-env`
3. Confirm the `nginx-ingress-controller` deployment exists:
``` console ```console
$ kubectl get pods -n ingress-nginx $ kubectl get pods -n ingress-nginx
NAME READY STATUS RESTARTS AGE NAME READY STATUS RESTARTS AGE
default-http-backend-66b447d9cf-rrlf9 1/1 Running 0 12s default-http-backend-66b447d9cf-rrlf9 1/1 Running 0 12s
nginx-ingress-controller-fdcdcd6dd-vvpgs 1/1 Running 0 11s nginx-ingress-controller-fdcdcd6dd-vvpgs 1/1 Running 0 11s
``` ```
### AWS #### AWS
In AWS we use an Elastic Load Balancer (ELB) to expose the NGINX Ingress controller behind a Service of `Type=LoadBalancer`. In AWS we use an Elastic Load Balancer (ELB) to expose the NGINX Ingress controller behind a Service of `Type=LoadBalancer`.
This setup requires to choose in which layer (L4 or L7) we want to configure the ELB: 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](https://aws.amazon.com/elasticloadbalancing/details/)
- [Layer 4](https://en.wikipedia.org/wiki/OSI_model#Layer_4:_Transport_Layer): use TCP as the listener protocol for ports 80 and 443. ##### Elastic Load Balancer - ELB
- [Layer 7](https://en.wikipedia.org/wiki/OSI_model#Layer_7:_Application_Layer): use HTTP as the listener protocol for port 80 and terminate TLS in the ELB
Patch the nginx ingress controller deployment to add the flag `--publish-service` This setup requires to choose in which layer (L4 or L7) we want to configure the ELB:
``` console - [Layer 4](https://en.wikipedia.org/wiki/OSI_model#Layer_4:_Transport_Layer): use TCP as the listener protocol for ports 80 and 443.
kubectl patch deployment -n ingress-nginx nginx-ingress-controller --type='json' \ - [Layer 7](https://en.wikipedia.org/wiki/OSI_model#Layer_7:_Application_Layer): use HTTP as the listener protocol for port 80 and terminate TLS in the ELB
--patch="$(curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/publish-service-patch.yaml)"
```
For L4: For L4:
``` console 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](#elb-idle-timeouts) 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/service-l4.yaml`
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/aws/service-l4.yaml
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/aws/patch-configmap-l4.yaml
```
For L7:
Change line of the file `provider/aws/service-l7.yaml` replacing the dummy id with a valid one `"arn:aws:acm:us-west-2:XXXXXXXX:certificate/XXXXXX-XXXXXXX-XXXXXXX-XXXXXXXX"`
Then execute: Then execute:
``` console ```console
kubectl apply -f provider/aws/service-l7.yaml kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/service-l4.yaml
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/aws/patch-configmap-l7.yaml kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/patch-configmap-l4.yaml
```
This example creates an ELB with just two listeners, one in port 80 and another in port 443
![Listeners](../docs/images/elb-l7-listener.png)
If the ingress controller uses RBAC run:
``` console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/patch-service-with-rbac.yaml
```
If not run:
``` console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/patch-service-without-rbac.yaml
``` ```
### GCE - GKE For L7:
Patch the nginx ingress controller deployment to add the flag `--publish-service` Change line of the file `provider/aws/service-l7.yaml` replacing the dummy id with a valid one `"arn:aws:acm:us-west-2:XXXXXXXX:certificate/XXXXXX-XXXXXXX-XXXXXXX-XXXXXXXX"`
``` console 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](#elb-idle-timeouts) 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/service-l7.yaml`
kubectl patch deployment -n ingress-nginx nginx-ingress-controller --type='json' \
--patch="$(curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/publish-service-patch.yaml)"
```
``` console Then execute:
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/gce-gke/service.yaml \
| kubectl apply -f -
```
If the ingress controller uses RBAC run:
``` console ```console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/patch-service-with-rbac.yaml kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/service-l7.yaml
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/patch-configmap-l7.yaml
``` ```
If not run: This example creates an ELB with just two listeners, one in port 80 and another in port 443
``` console ![Listeners](https://github.com/kubernetes/ingress-nginx/raw/master/docs/images/elb-l7-listener.png)
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/patch-service-without-rbac.yaml
```
**Important Note:** proxy protocol is not supported in GCE/GKE ##### 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](http://nginx.org/en/docs/http/ngx_http_core_module.html#keepalive_timeout) that is configured for NGINX. By default NGINX `keepalive_timeout` is set to `75s`.
### Azure The default ELB idle timeout will work for most scenarios, unless the NGINX [keepalive_timeout](http://nginx.org/en/docs/http/ngx_http_core_module.html#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.
Patch the nginx ingress controller deployment to add the flag `--publish-service` _Please Note: An idle timeout of `3600s` is recommended when using WebSockets._
``` console More information with regards to idle timeouts for your Load Balancer can be found in the [official AWS documentation](https://docs.aws.amazon.com/elasticloadbalancing/latest/classic/config-idle-timeout.html).
kubectl patch deployment -n ingress-nginx nginx-ingress-controller --type='json' \
--patch="$(curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/publish-service-patch.yaml)"
```
``` console ##### Network Load Balancer (NLB)
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/azure/service.yaml \
| kubectl apply -f -
```
If the ingress controller uses RBAC run: This type of load balancer is supported since v1.10.0 as an ALPHA feature.
``` console ```console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/patch-service-with-rbac.yaml kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/aws/service-nlb.yaml
``` ```
If not run: #### GCE-GKE
``` console ```console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/patch-service-without-rbac.yaml 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 **Important Note:** proxy protocol is not supported in GCE/GKE
### Baremetal #### Azure
Using [NodePort](https://kubernetes.io/docs/concepts/services-networking/service/#type-nodeport): ```console
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/cloud-generic.yaml
``` console
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/baremetal/service-nodeport.yaml \
| kubectl apply -f -
``` ```
Using Helm #### Bare-metal
----------
NGINX Ingress controller can be installed via [Helm](https://helm.sh/) using the chart [stable/nginx](https://github.com/kubernetes/charts/tree/master/stable/nginx-ingress) from the official charts repository. Using [NodePort](https://kubernetes.io/docs/concepts/services-networking/service/#type-nodeport):
To install the chart with the release name `my-nginx`:
``` console ```console
helm install stable/nginx-ingress --name my-nginx kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/static/provider/baremetal/service-nodeport.yaml
``` ```
If the kubernetes cluster has RBAC enabled, then run: !!! tip
For extended notes regarding deployments on bare-metal, see [Bare-metal considerations](./baremetal.md).
``` console
helm install stable/nginx-ingress --name my-nginx --set rbac.create=true
```
Verify installation ### Verify installation
-------------------
To check if the ingress controller pods have started, run the following command: To check if the ingress controller pods have started, run the following command:
``` console ```console
kubectl get pods --all-namespaces -l app=ingress-nginx --watch 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`. 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. Now, you are ready to create your first ingress.
Detect installed version ### Detect installed version
------------------------
To detect which version of the ingress controller is running, exec into the pod and run `nginx-ingress-controller version` command. To detect which version of the ingress controller is running, exec into the pod and run `nginx-ingress-controller version` command.
``` console ```console
POD_NAMESPACE=ingress-nginx POD_NAMESPACE=ingress-nginx
POD_NAME=$(kubectl get pods -n $POD_NAMESPACE -l app=ingress-nginx -o jsonpath={.items[0].metadata.name}) POD_NAME=$(kubectl get pods -n $POD_NAMESPACE -l app.kubernetes.io/name=ingress-nginx -o jsonpath='{.items[0].metadata.name}')
kubectl exec -it $POD_NAME -n $POD_NAMESPACE -- /nginx-ingress-controller --version kubectl exec -it $POD_NAME -n $POD_NAMESPACE -- /nginx-ingress-controller --version
``` ```
Deploying the config-map ## Using Helm
------------------------
A config map can be used to configure system components for the nginx-controller. In order to begin using a config-map
make sure it has been created and is being used in the deployment.
It is created as seen in the [Mandatory Commands](#mandatory-commands) section above. NGINX Ingress controller can be installed via [Helm](https://helm.sh/) using the chart [stable/nginx-ingress](https://github.com/kubernetes/charts/tree/master/stable/nginx-ingress) from the official charts repository.
To install the chart with the release name `my-nginx`:
``` console ```console
curl https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/configmap.yaml \ helm install stable/nginx-ingress --name my-nginx
| kubectl apply -f -
``` ```
and is setup to be used in the deployment [without-rbac](without-rbac.yaml) or [with-rbac](with-rbac.yaml) with the following line: If the kubernetes cluster has RBAC enabled, then run:
``` yaml ```console
- --configmap=$(POD_NAMESPACE)/nginx-configuration helm install stable/nginx-ingress --name my-nginx --set rbac.create=true
``` ```
For information on using the config-map, see its [user-guide](../docs/user-guide/configmap.md). Detect installed version:
```console
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
```
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment