The admission controllers documentation introduces how to use standard, plugin-style admission controllers. However, plugin admission controllers are not flexible enough for all use cases, due to the following:
1.7 introduces two alpha features, Initializers and External Admission Webhooks, that address these limitations. These features allow admission controllers to be developed out-of-tree and configured at runtime.
This page describes how to use Initializers and External Admission Webhooks.
Initializer has two meanings:
A list of pending pre-initialization tasks, stored in every object’s metadata (e.g., “AddMyCorporatePolicySidecar”).
A user customized controller, which actually perform those tasks. The name of the task corresponds to the controller which performs the task. For clarity, we call them initializer controllers in this page.
Once the controller has performed its assigned task, it removes its name from
the list. For example, it may send a PATCH that inserts a container in a pod and
also removes its name from metadata.initializers.pending
. Initializers may make
mutations to objects.
Objects which have a non-empty initializer list are considered uninitialized,
and are not visible in the API unless specifically requested by using the query parameter,
?includeUninitialized=true
.
Initializers are useful for admins to force policies (e.g., the AlwaysPullImages admission controller), or to inject defaults (e.g., the DefaultStorageClass admission controller), etc.
Note: If your use case does not involve mutating objects, consider using external admission webhooks, as they have better performance.
When an object is POSTed, it is checked against all existing
initializerConfiguration
objects (explained below). For all that it matches,
all spec.initializers[].name
s are appended to the new object’s
metadata.initializers.pending
field.
An initializer controller should list and watch for uninitialized objects, by
using the query parameter ?includeUninitialized=true
. If using client-go, just
set
listOptions.includeUninitialized
to true.
For the observed uninitialized objects, an initializer controller should first
check if its name matches metadata.initializers.pending[0]
. If so, it should then
perform its assigned task and remove its name from the list.
Initializers is an alpha feature, so it is disabled by default. To turn it on, you need to:
Include “Initializers” in the --admission-control
flag when starting
kube-apiserver
. If you have multiple kube-apiserver
replicas, all should
have the same flag setting.
Enable the dynamic admission controller registration API by adding
admissionregistration.k8s.io/v1alpha1
to the --runtime-config
flag passed
to kube-apiserver
, e.g.
--runtime-config=admissionregistration.k8s.io/v1alpha1
. Again, all replicas
should have the same flag setting.
You should deploy an initializer controller via the deployment API.
You can configure what initializers are enabled and what resources are subject
to the initializers by creating initializerConfiguration
resources.
You should first deploy the initializer controller and make sure that it is
working properly before creating the initializerConfiguration
. Otherwise, any
newly created resources will be stuck in an uninitialized state.
The following is an example initializerConfiguration
:
apiVersion: admissionregistration.k8s.io/v1alpha1
kind: InitializerConfiguration
metadata:
name: example-config
initializers:
# the name needs to be fully qualified, i.e., containing at least two "."
- name: podimage.example.com
rules:
# apiGroups, apiVersion, resources all support wildcard "*".
# "*" cannot be mixed with non-wildcard.
- apiGroups:
- ""
apiVersions:
- v1
resources:
- pods
After you create the initializerConfiguration
, the system will take a few
seconds to honor the new configuration. Then, "podimage.example.com"
will be
appended to the metadata.initializers.pending
field of newly created pods. You
should already have a ready “podimage” initializer controller that handles pods
whose metadata.initializers.pending[0].name="podimage.example.com"
. Otherwise
the pods will stuck uninitialized.
Make sure that all expansions of the <apiGroup, apiVersions, resources>
tuple
in a rule
are valid. If they are not, separate them in different rules
.
External admission webhooks are HTTP callbacks that are intended to receive admission requests and do something with them. What an external admission webhook does is up to you, but there is an interface that it must adhere to so that it responds with whether or not the admission request should be allowed.
Unlike initializers or the plugin-style admission controllers, external admission webhooks are not allowed to mutate the admission request in any way.
Because admission is a high security operation, the external admission webhooks must support TLS.
A simple example use case for an external admission webhook is to do semantic validation
of Kubernetes resources. Suppose that your infrastructure requires that all Pod
resources have a common set of labels, and you do not want any Pod
to be
persisted to Kubernetes if those needs are not met. You could write your
external admission webhook to do this validation and respond accordingly.
Whenever a request comes in, the GenericAdmissionWebhook
admission plugin will
get the list of interested external admission webhooks from
externalAdmissionHookConfiguration
objects (explained below) and call them in
parallel. If all of the external admission webhooks approve the admission
request, the admission chain continues. If any of the external admission
webhooks deny the admission request, the admission request will be denied, and
the reason for doing so will be based on the first external admission webhook
denial reason. This means if there is more than one external admission webhook
that denied the admission request, only the first will be returned to the
user. If there is an error encountered when calling an external admission
webhook, that request is ignored and will not be used to approve/deny the
admission request.
Note: The admission chain depends solely on the order of the
--admission-control
option passed to kube-apiserver
.
External Admission Webhooks is an alpha feature, so it is disabled by default. To turn it on, you need to
Include “GenericAdmissionWebhook” in the --admission-control
flag when
starting the apiserver. If you have multiple kube-apiserver
replicas, all
should have the same flag setting.
Enable the dynamic admission controller registration API by adding
admissionregistration.k8s.io/v1alpha1
to the --runtime-config
flag passed
to kube-apiserver
, e.g.
--runtime-config=admissionregistration.k8s.io/v1alpha1
. Again, all replicas
should have the same flag setting.
See caesarxuchao/example-webhook-admission-controller for an example webhook admission controller.
The communication between the webhook admission controller and the apiserver, or
more precisely, the GenericAdmissionWebhook admission controller, needs to be
TLS secured. You need to generate a CA cert and use it to sign the server cert
used by your webhook admission controller. The pem formatted CA cert is supplied
to the apiserver via the dynamic registration API
externaladmissionhookconfigurations.clientConfig.caBundle
.
For each request received by the apiserver, the GenericAdmissionWebhook
admission controller sends an
admissionReview
to the relevant webhook admission controller. The webhook admission controller
gathers information like object
, oldobject
, and userInfo
, from
admissionReview.spec
, sends back a response with the body also being the
admissionReview
, whose status
field is filled with the admission decision.
See caesarxuchao/example-webhook-admission-controller deployment for an example deployment.
The webhook admission controller should be deployed via the deployment API. You also need to create a service as the front-end of the deployment.
You can configure what webhook admission controllers are enabled and what resources are subject to the admission controller via creating externaladmissionhookconfigurations.
We suggest that you first deploy the webhook admission controller and make sure it is working properly before creating the externaladmissionhookconfigurations. Otherwise, depending whether the webhook is configured as fail open or fail closed, operations will be unconditionally accepted or rejected.
The following is an example externaladmissionhookconfiguration
:
apiVersion: admissionregistration.k8s.io/v1alpha1
kind: ExternalAdmissionHookConfiguration
metadata:
name: example-config
externalAdmissionHooks:
- name: pod-image.k8s.io
rules:
- apiGroups:
- ""
apiVersions:
- v1
operations:
- CREATE
resources:
- pods
failurePolicy: Ignore
clientConfig:
caBundle: <pem encoded ca cert that signs the server cert used by the webhook>
service:
name: <name of the front-end service>
namespace: <namespace of the front-end service>
For a request received by the apiserver, if the request matches any of the
rules
of an externalAdmissionHook
, the GenericAdmissionWebhook
admission
controller will send an admissionReview
request to the externalAdmissionHook
to ask for admission decision.
The rule
is similar to the rule
in initializerConfiguration
, with two
differences:
The addition of the operations
field, specifying what operations the webhook
is interested in;
The resources
field accepts subresources in the form or resource/subresource.
Make sure that all expansions of the <apiGroup, apiVersions,resources>
tuple
in a rule
are valid. If they are not, separate them to different rules
.
You can also specify the failurePolicy
. In 1.7, the system supports Ignore
and Fail
policies, meaning that upon a communication error with the webhook
admission controller, the GenericAdmissionWebhook
can admit or reject the
operation based on the configured policy.
After you create the initializerConfiguration
, the system will take a few
seconds to honor the new configuration.