The Ambassador Edge Stack is a self-service, comprehensive edge stack that is Kubernetes-native and built on Envoy Proxy.

$ helm repo add datawire https://getambassador.io
$ helm install ambassador datawire/ambassador

This chart bootstraps an Ambassador deployment on a Kubernetes cluster using the Helm package manager.

• Kubernetes 1.11+

helm repo add datawire https://getambassador.io

To install the chart with the release name my-release:

$ kubectl create namespace ambassador
$ helm install my-release datawire/ambassador -n ambassador

The command deploys Ambassador Edge Stack in the ambassador namespace on the Kubernetes cluster in the default configuration.

It is recommended to use the ambassador namespace for easy upgrades.

The configuration section lists the parameters that can be configured during installation.

This chart defaults to installing The Ambassador Edge Stack with all of its configuration objects.

• A Redis instance
• AuthService resource for enabling authentication
• RateLimitService resource for enabling rate limiting
• Mappings for internal request routing

If installing alongside another deployment of Ambassador, some of these resources can cause configuration errors since only one AuthService or RateLimitService can be configured at a time.

If you already have one of these resources configured in your cluster, please see the configuration section below for information on how to disable them in the chart.

This chart can still be used to install Ambassador OSS.

To install OSS, change the image to use the OSS image and set enableAES: false to skip the install of any AES resources.

To uninstall/delete the my-release deployment:

$ helm uninstall my-release

The command removes all the Kubernetes components associated with the chart and deletes the release.

Notable chart changes are listed in the CHANGELOG

The following tables lists the configurable parameters of the Ambassador chart and their default values.

NOTE: Make sure the configured service.http.targetPort and service.https.targetPort ports match your Ambassador Module's service_port and redirect_cleartext_from configurations.

The Ambassador Edge Stack provides a comprehensive, self-service edge stack in the Kubernetes cluster with a decentralized deployment model and a declarative paradigm.

By default, this chart will install the latest image of The Ambassador Edge Stack which will replace your existing deployment of Ambassador with no changes to functionality.

This helm chart includes the creation of the core CRDs Ambassador uses for configuration.

The crds flags (Helm 2 only) let you configure how a release manages crds.

• crds.create Can only be set on your first/master Ambassador release.
• crds.enabled Should be set on all releases using Ambassador CRDs
• crds.keep Configures if the CRDs are deleted when the master release is purged. This value is only checked for the master release and can be set to any value on secondary releases.

Ambassador takes security very seriously. For this reason, the YAML installation will default with a couple of basic security policies in place.

The security field of the values.yaml file configures these default policies and replaces the securityContext field used earlier.

The defaults will configure the pod to run as a non-root user and prohibit privilege escalation and outline a PodSecurityPolicy to ensure these conditions are met.

security:
  # Security Context for all containers in the pod.
  # https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.18/#podsecuritycontext-v1-core
  podSecurityContext:
    runAsUser: 8888
  # Security Context for the Ambassador container specifically
  # https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.18/#securitycontext-v1-core
  containerSecurityContext:
    allowPrivilegeEscalation: false
  # A basic PodSecurityPolicy to ensure Ambassador is running with appropriate security permissions
  # https://kubernetes.io/docs/concepts/policy/pod-security-policy/
  #
  # A set of reasonable defaults is outlined below. This is not created by default as it should only
  # be created by a one Release. If you want to use the PodSecurityPolicy in the chart, create it in
  # the "master" Release and then leave it unset in all others. Set the `rbac.podSecurityPolicies` 
  # in all non-"master" Releases.
  podSecurityPolicy: {}
    # # Add AppArmor and Seccomp annotations
    # # https://kubernetes.io/docs/concepts/policy/pod-security-policy/#apparmor
    # annotations:
    # spec:
    #   seLinux:
    #     rule: RunAsAny
    #   supplementalGroups:
    #     rule: 'MustRunAs'
    #     ranges:
    #       # Forbid adding the root group.
    #       - min: 1
    #         max: 65535
    #   fsGroup:
    #     rule: 'MustRunAs'
    #     ranges:
    #       # Forbid adding the root group.
    #       - min: 1
    #         max: 65535
    #   privileged: false
    #   allowPrivilegeEscalation: false
    #   runAsUser:
    #     rule: MustRunAsNonRoot

Ambassador is configured using Kubernetes Custom Resource Definitions (CRDs). If you are unable to use CRDs, Ambassador can also be configured using annotations on services. The service.annotations section of the values file contains commented out examples of Ambassador Module and a global TLSContext configurations which are typically created in the Ambassador service.

If you intend to use service.annotations, remember to include the getambassador.io/config annotation key as above.

Using the Prometheus Exporter has been deprecated and is no longer recommended. You can now use metrics.serviceMonitor.enabled to create a ServiceMonitor from the chart if the Prometheus Operator has been installed on your cluster.

Please see Ambassador's monitoring with Prometheus docs for more information on using the /metrics endpoint for metrics collection.

Specify each parameter using the --set key=value[,key=value] argument to helm install. For example,

$ helm install --wait my-release \
    --set adminService.type=NodePort \
    datawire/ambassador

Alternatively, a YAML file that specifies the values for the above parameters can be provided while installing the chart. For example,

$ helm install --wait my-release -f values.yaml datawire/ambassador

Introduces Ambassador Edge Stack being installed by default.

Ambassador Pro support has been removed in 6.0.0. Please upgrade to the Ambassador Edge Stack.

Note If upgrading an existing helm 2 installation no action is needed, previously installed CRDs will not be modified.

• Helm 3 support for CRDs was added. Specifically, the CRD templates were moved to non-templated files in the /crds directory, and to keep Helm 2 support they are globbed from there by /templates/crds.yaml. However, because Helm 3 CRDs are not templated, the labels for new installations have necessarily changed

The 4.0.0 chart contains a number of changes to the way Ambassador Pro is installed.

• Introduces the performance tuned and certified build of open source Ambassador, Ambassador core
• The license key is now stored and read from a Kubernetes secret by default
• Added .Values.pro.licenseKey.secret.enabled .Values.pro.licenseKey.secret.create fields to allow multiple releases in the same namespace to use the same license key secret.
• Introduces the ability to configure resource limits for both Ambassador Pro and it's redis instance
• Introduces the ability to configure additional AuthService options (see AuthService documentation)
• The ambassador-pro-auth AuthService and ambassador-pro-ratelimit RateLimitService and now created as CRDs when .Values.crds.enabled: true
• Fixed misnamed selector for redis instance that failed in an edge case
• Exposes annotations for redis deployment and service

The value of .Values.pro.image.tag has been shortened to assume amb-sidecar (and amb-core for Ambassador core) values.yaml

<3.0.0>
  image:
    repository: quay.io/datawire/ambassador_pro
-    tag: amb-sidecar-0.6.0

<4.0.0+>
  image:
    repository: quay.io/datawire/ambassador_pro
+    tag: 0.7.0

Method for creating a Kubernetes secret to hold the license key has been changed

values.yaml

<3.0.0>
-    secret: false
<4.0.0>
+    secret:
+      enabled: true
+      create: true

The way ports are assigned has been changed for a more dynamic method.

Now, instead of setting the port assignments for only the http and https, any port can be open on the load balancer using a list like you would in a standard Kubernetes YAML manifest.

pre-3.0.0

service:
  http:
    enabled: true
    port: 80
    targetPort: 8080
  https:
    enabled: true
    port: 443
    targetPort: 8443

3.0.0

service:
  ports:
  - name: http
    port: 80
    targetPort: 8080
  - name: https
    port: 443
    targetPort: 8443

This change has also replaced the .additionalTCPPorts configuration. Additional TCP ports can be created the same as the http and https ports above.

The below Ambassador Module annotation is no longer being applied by default.

getambassador.io/config: |
  ---
  apiVersion: ambassador/v1
  kind: Module
  name: ambassador
  config:
    service_port: 8080

This was causing confusion with the service_port being hard-coded when enabling TLS termination in Ambassador.

Ambassador has been listening on port 8080 for HTTP and 8443 for HTTPS by default since version 0.60.0 (chart version 2.2.0).

A ClusterRole and ClusterRoleBinding named {{release name}}-crd will be created to watch for the Ambassador Custom Resource Definitions. This will be created regardless of the value of scope.singleNamespace since CRDs are created the cluster scope.

rbac.namespaced has been removed. For namespaced RBAC, set scope.singleNamespace: true and rbac.enabled: true.

crds.enabled will indicate that you are using CRDs and will create the rbac resources regardless of the value of crds.create. This allows for multiple deployments to use the CRDs.

ambassador.id has been removed in favor of setting it via an environment variable in env. AMBASSADOR_ID defaults to default if not set in the environment. This is mainly used for running multiple Ambassadors in the same cluster.

Chart now runs ambassador as non-root by default, so you might need to update your ambassador module config to match this.

Timings values have been removed in favor of setting the env variables using `env´

Service ports values have changed names and target ports have new defaults.

Pre version 0.50.0 ambassador was using socat and required a sidecar to export statsd metrics. In 0.50.0 ambassador no longer uses socat and doesn't need a sidecar anymore to export its statsd metrics. Statsd metrics are disabled by default and can be enabled by setting environment STATSD_ENABLED, this will (in 0.50) send metrics to a service named statsd-sink, if you want to send it to another service or namespace it can be changed by setting STATSD_HOST

If you are using prometheus the chart allows you to enable a sidecar which can export to prometheus see the prometheusExporter values.