1. For Nginx Ingress controllers used for non-critical services, the easiest method is to [uninstall](#uninstall-chart) the old release and [install](#install-chart) the new one
1. For critical services in production that require zero-downtime, you will want to:
1. [Install](#install-chart) a second Ingress controller
1. Redirect your DNS traffic from the old controller to the new controller
1. Log traffic from both controllers during this changeover
1. [Uninstall](#uninstall-chart) the old controller once traffic has fully drained from it
Note that there are some different and upgraded configurations between the two charts, described by Rimas Mocevicius from JFrog in the "Upgrading to ingress-nginx Helm chart" section of [Migrating from Helm chart nginx-ingress to ingress-nginx](https://rimusz.net/migrating-to-ingress-nginx). As the `ingress-nginx/ingress-nginx` chart continues to update, you will want to check current differences by running [helm configuration](#configuration) commands on both charts.
See [Customizing the Chart Before Installing](https://helm.sh/docs/intro/using_helm/#customizing-the-chart-before-installing). To see all configurable options with detailed comments, visit the chart's [values.yaml](./values.yaml), or run these configuration commands:
You can add Prometheus annotations to the metrics service using `controller.metrics.service.annotations`.
Alternatively, if you use the Prometheus Operator, you can enable ServiceMonitor creation using `controller.metrics.serviceMonitor.enabled`. And set `controller.metrics.serviceMonitor.additionalLabels.release="prometheus"`. "release=prometheus" should match the label configured in the prometheus servicemonitor ( see `kubectl get servicemonitor prometheus-kube-prom-prometheus -oyaml -n prometheus`)
Previous versions of this chart had a `controller.stats.*` configuration block, which is now obsolete due to the following changes in Ingress-Nginx Controller:
- In [0.16.1](https://github.com/kubernetes/ingress-nginx/blob/main/Changelog.md#0161), the vts (virtual host traffic status) dashboard was removed
- In [0.23.0](https://github.com/kubernetes/ingress-nginx/blob/main/Changelog.md#0230), the status page at port 18080 is now a unix socket webserver only available at localhost.
You can use `curl --unix-socket /tmp/nginx-status-server.sock http://localhost/nginx_status` inside the controller container to access it locally, or use the snippet from [nginx-ingress changelog](https://github.com/kubernetes/ingress-nginx/blob/main/Changelog.md#0230) to re-enable the http server
Annotate the controller as shown in the [nginx-ingress l7 patch](https://github.com/kubernetes/ingress-nginx/blob/ab3a789caae65eec4ad6e3b46b19750b481b6bce/deploy/aws/l7/service-l7.yaml):
This setup is useful when you need both external and internal load balancers but don't want to have multiple ingress controllers and multiple ingress objects per application.
By default, the ingress object will point to the external load balancer address, but if correctly configured, you can make use of the internal one if the URL you are looking up resolves to the internal load balancer's URL.
You'll need to set both the following values:
`controller.service.internal.enabled`
`controller.service.internal.annotations`
If one of them is missing the internal load balancer will not be deployed. Example you may have `controller.service.internal.enabled=true` but no annotations set, in this case no action will be taken.
`controller.service.internal.annotations` varies with the cloud service you're using.
An use case for this scenario is having a split-view DNS setup where the public zone CNAME records point to the external balancer URL while the private zone CNAME records point to the internal balancer URL. This way, you only need one ingress kubernetes object.
With nginx-ingress-controller version 0.25+, the Ingress-Nginx Controller pod exposes an endpoint that will integrate with the `validatingwebhookconfiguration` Kubernetes feature to prevent bad ingress from being added to the cluster.
A validating and configuration requires the endpoint to which the request is sent to use TLS. It is possible to set up custom certificates to do this, but in most cases, a self-signed certificate is enough. The setup of this component requires some more complex orchestration when using helm. The steps are created to be idempotent and to allow turning the feature on and off without running into helm quirks.
1. A pre-install hook provisions a certificate into the same namespace using a format compatible with provisioning using end user certificates. If the certificate already exists, the hook exits.
3. Validating and Mutating webhook configurations are created in the cluster.
4. A post-install hook reads the CA from the secret created by step 1 and patches the Validating and Mutating webhook configurations. This process will allow a custom CA provisioned by some other process to also be patched into the webhook configurations. The chosen failure policy is also patched into the webhook configurations
#### Alternatives
It should be possible to use [cert-manager/cert-manager](https://github.com/cert-manager/cert-manager) if a more complete solution is required.
You can enable automatic self-signed TLS certificate provisioning via cert-manager by setting the `controller.admissionWebhooks.certManager.enabled` value to true.
Error: UPGRADE FAILED: Service "?????-controller" is invalid: spec.clusterIP: Invalid value: "": field is immutable
```
Detail of how and why are in [this issue](https://github.com/helm/charts/pull/13646) but to resolve this you can set `xxxx.service.omitClusterIP` to `true` where `xxxx` is the service referenced in the error.
As of version `1.26.0` of this chart, by simply not providing any clusterIP value, `invalid: spec.clusterIP: Invalid value: "": field is immutable` will no longer occur since `clusterIP: ""` will not be rendered.
| controller.addHeaders | object | `{}` | Will add custom headers before sending response traffic to the client according to: https://kubernetes.github.io/ingress-nginx/user-guide/nginx-configuration/configmap/#add-headers |
| controller.allowSnippetAnnotations | bool | `true` | This configuration defines if Ingress Controller should allow users to set their own *-snippet annotations, otherwise this is forbidden / dropped when users add those annotations. Global snippets in ConfigMap are still respected |
| controller.dnsConfig | object | `{}` | Optionally customize the pod dnsConfig. |
| controller.dnsPolicy | string | `"ClusterFirst"` | Optionally change this to ClusterFirstWithHostNet in case you have 'hostNetwork: true'. By default, while using host network, name resolution uses the host's DNS. If you wish nginx-controller to keep resolving names inside the k8s network, use ClusterFirstWithHostNet. |
| controller.electionID | string | `""` | Election ID to use for status update, by default it uses the controller name combined with a suffix of 'leader' |
| controller.enableTopologyAwareRouting | bool | `false` | This configuration enables Topology Aware Routing feature, used together with service annotation service.kubernetes.io/topology-aware-hints="auto" Defaults to false |
| controller.extraArgs | object | `{}` | Additional command line arguments to pass to Ingress-Nginx Controller E.g. to specify the default SSL certificate you can use |
| controller.extraContainers | list | `[]` | Additional containers to be added to the controller pod. See https://github.com/lemonldap-ng-controller/lemonldap-ng-controller as example. |
| controller.extraEnvs | list | `[]` | Additional environment variables to set |
| controller.extraInitContainers | list | `[]` | Containers, which are run before the app containers are started. |
| controller.extraModules | list | `[]` | Modules, which are mounted into the core nginx image. See values.yaml for a sample to add opentelemetry module |
| controller.healthCheckHost | string | `""` | Address to bind the health check endpoint. It is better to set this option to the internal node address if the Ingress-Nginx Controller is running in the `hostNetwork: true` mode. |
| controller.healthCheckPath | string | `"/healthz"` | Path of the health check endpoint. All requests received on the port defined by the healthz-port parameter are forwarded internally to this path. |
| controller.hostNetwork | bool | `false` | Required for use with CNI based kubernetes installations (such as ones set up by kubeadm), since CNI and hostport don't mix yet. Can be deprecated once https://github.com/kubernetes/kubernetes/issues/23920 is merged |
| controller.hostPort.enabled | bool | `false` | Enable 'hostPort' or not |
| controller.hostPort.ports.http | int | `80` | 'hostPort' http port |
| controller.hostPort.ports.https | int | `443` | 'hostPort' https port |
| controller.hostname | object | `{}` | Optionally customize the pod hostname. |
| controller.ingressClass | string | `"nginx"` | For backwards compatibility with ingress.class annotation, use ingressClass. Algorithm is as follows, first ingressClassName is considered, if not present, controller looks for ingress.class annotation |
| controller.ingressClassResource.controllerValue | string | `"k8s.io/ingress-nginx"` | Controller-value of the controller that is processing this ingressClass |
| controller.ingressClassResource.default | bool | `false` | Is this the default ingressClass for the cluster |
| controller.ingressClassResource.enabled | bool | `true` | Is this ingressClass enabled or not |
| controller.ingressClassResource.name | string | `"nginx"` | Name of the ingressClass |
| controller.ingressClassResource.parameters | object | `{}` | Parameters is a link to a custom resource containing additional configuration for the controller. This is optional if the controller does not require extra parameters. |
| controller.labels | object | `{}` | Labels to be added to the controller Deployment or DaemonSet and other resources that do not have option to specify labels # |
| controller.lifecycle | object | `{"preStop":{"exec":{"command":["/wait-shutdown"]}}}` | Improve connection draining when ingress controller pod is deleted using a lifecycle hook: With this new hook, we increased the default terminationGracePeriodSeconds from 30 seconds to 300, allowing the draining of connections up to five minutes. If the active connections end before that, the pod will terminate gracefully at that time. To effectively take advantage of this feature, the Configmap feature worker-shutdown-timeout new value is 240s instead of 10s. # |
| controller.metrics.service.externalIPs | list | `[]` | List of IP addresses at which the stats-exporter service is available # Ref: https://kubernetes.io/docs/concepts/services-networking/service/#external-ips # |
| controller.minAvailable | int | `1` | Minimum available pods set in PodDisruptionBudget. Define either 'minAvailable' or 'maxUnavailable', never both. |
| controller.proxySetHeaders | object | `{}` | Will add custom headers before sending traffic to backends according to https://github.com/kubernetes/ingress-nginx/tree/main/docs/examples/customization/custom-headers |
| controller.publishService | object | `{"enabled":true,"pathOverride":""}` | Allows customization of the source of the IP address or FQDN to report in the ingress status field. By default, it reads the information provided by the service. If disable, the status field reports the IP address of the node or nodes where an ingress controller pod is running. |
| controller.publishService.enabled | bool | `true` | Enable 'publishService' or not |
| controller.publishService.pathOverride | string | `""` | Allows overriding of the publish service to bind to Must be <namespace>/<service_name> |
| controller.readinessProbe.failureThreshold | int | `3` | |
| controller.reportNodeInternalIp | bool | `false` | Bare-metal considerations via the host network https://kubernetes.github.io/ingress-nginx/deploy/baremetal/#via-the-host-network Ingress status was blank because there is no Service exposing the Ingress-Nginx Controller in a configuration using the host network, the default --publish-service flag used in standard cloud setups does not apply |
| controller.scope.enabled | bool | `false` | Enable 'scope' or not |
| controller.scope.namespace | string | `""` | Namespace to limit the controller to; defaults to $(POD_NAMESPACE) |
| controller.scope.namespaceSelector | string | `""` | When scope.enabled == false, instead of watching all namespaces, we watching namespaces whose labels only match with namespaceSelector. Format like foo=bar. Defaults to empty, means watching all namespaces. |
| controller.service.appProtocol | bool | `true` | If enabled is adding an appProtocol option for Kubernetes service. An appProtocol field replacing annotations that were using for setting a backend protocol. Here is an example for AWS: service.beta.kubernetes.io/aws-load-balancer-backend-protocol: http It allows choosing the protocol for each backend specified in the Kubernetes service. See the following GitHub issue for more details about the purpose: https://github.com/kubernetes/kubernetes/issues/40244 Will be ignored for Kubernetes versions older than 1.20 # |
| controller.service.externalIPs | list | `[]` | List of IP addresses at which the controller services are available # Ref: https://kubernetes.io/docs/concepts/services-networking/service/#external-ips # |
| controller.service.internal.annotations | object | `{}` | Annotations are mandatory for the load balancer to come up. Varies with the cloud service. |
| controller.service.internal.enabled | bool | `false` | Enables an additional internal load balancer (besides the external one). |
| controller.service.internal.loadBalancerSourceRanges | list | `[]` | Restrict access For LoadBalancer service. Defaults to 0.0.0.0/0. |
| controller.service.ipFamilies | list | `["IPv4"]` | List of IP families (e.g. IPv4, IPv6) assigned to the service. This field is usually assigned automatically based on cluster configuration and the ipFamilyPolicy field. # Ref: https://kubernetes.io/docs/concepts/services-networking/dual-stack/ |
| controller.service.ipFamilyPolicy | string | `"SingleStack"` | Represents the dual-stack-ness requested or required by this Service. Possible values are SingleStack, PreferDualStack or RequireDualStack. The ipFamilies and clusterIPs fields depend on the value of this field. # Ref: https://kubernetes.io/docs/concepts/services-networking/dual-stack/ |
| controller.service.loadBalancerIP | string | `""` | Used by cloud providers to connect the resulting `LoadBalancer` to a pre-existing static IP according to https://kubernetes.io/docs/concepts/services-networking/service/#loadbalancer |
| controller.sysctls | object | `{}` | See https://kubernetes.io/docs/tasks/administer-cluster/sysctl-cluster/ for notes on enabling and using sysctls |
| controller.tcp.annotations | object | `{}` | Annotations to be added to the tcp config configmap |
| controller.tcp.configMapNamespace | string | `""` | Allows customization of the tcp-services-configmap; defaults to $(POD_NAMESPACE) |
| controller.terminationGracePeriodSeconds | int | `300` | `terminationGracePeriodSeconds` to avoid killing pods before we are ready # wait up to five minutes for the drain of connections # |
| controller.tolerations | list | `[]` | Node tolerations for server scheduling to nodes with taints # Ref: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/ # |
| controller.topologySpreadConstraints | list | `[]` | Topology spread constraints rely on node labels to identify the topology domain(s) that each Node is in. # Ref: https://kubernetes.io/docs/concepts/workloads/pods/pod-topology-spread-constraints/ # |
| controller.watchIngressWithoutClass | bool | `false` | Process Ingress objects without ingressClass annotation/ingressClassName field Overrides value for --watch-ingress-without-class flag of the controller binary Defaults to false |
| defaultBackend.containerSecurityContext | object | `{}` | Security Context policies for controller main container. See https://kubernetes.io/docs/tasks/administer-cluster/sysctl-cluster/ for notes on enabling and using sysctls # |
| defaultBackend.podSecurityContext | object | `{}` | Security Context policies for controller pods See https://kubernetes.io/docs/tasks/administer-cluster/sysctl-cluster/ for notes on enabling and using sysctls # |
| defaultBackend.service.externalIPs | list | `[]` | List of IP addresses at which the default backend service is available # Ref: https://kubernetes.io/docs/concepts/services-networking/service/#external-ips # |
| defaultBackend.tolerations | list | `[]` | Node tolerations for server scheduling to nodes with taints # Ref: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/ # |
| portNamePrefix | string | `""` | Prefix for TCP and UDP ports names in ingress controller service # Some cloud providers, like Yandex Cloud may have a requirements for a port name regex to support cloud load balancer integration |
