Identifies the issuer that issued the JWT. See issuer Usually a URL or an email address.

Example: https://securetoken.google.com Example: 1234567-compute@developer.gserviceaccount.com

The list of JWT audiences. that are allowed to access. A JWT containing any of these audiences will be accepted.

The service name will be accepted if audiences is empty.

Example:

audiences:
- bookstore_android.apps.googleusercontent.com
  bookstore_web.apps.googleusercontent.com

URL of the provider’s public key set to validate signature of the JWT. See OpenID Discovery.

Optional if the key set document can either (a) be retrieved from OpenID Discovery of the issuer or (b) inferred from the email domain of the issuer (e.g. a Google service account).

Example: https://www.googleapis.com/oauth2/v1/certs

Note: Only one of jwks_uri and jwks should be used.

JSON Web Key Set of public keys to validate signature of the JWT. See https://auth0.com/docs/jwks.

Note: Only one of jwks_uri and jwks should be used.

JWT is sent in a request header. header represents the header name.

For example, if header=x-goog-iap-jwt-assertion, the header format will be x-goog-iap-jwt-assertion: <JWT>.

JWT is sent in a query parameter. query represents the query parameter name.

For example, query=jwt_token.

List of trigger rules to decide if this JWT should be used to validate the request. The JWT validation happens if any one of the rules matched. If the list is not empty and none of the rules matched, authentication will skip the JWT validation. Leave this empty to always trigger the JWT validation.

List of paths to be excluded from the request. The rule is satisfied if request path does not match to any of the path in this list.

List of paths that the request must include. If the list is not empty, the rule is satisfied if request path matches at least one of the path in the list. If the list is empty, the rule is ignored, in other words the rule is always satisfied.

WILL BE DEPRECATED, if set, will translates to TLS_PERMISSIVE mode. Set this flag to true to allow regular TLS (i.e without client x509 certificate). If request carries client certificate, identity will be extracted and used (set to peer identity). Otherwise, peer identity will be left unset. When the flag is false (default), request must have client certificate.

Defines the mode of mTLS authentication.

Client cert must be presented, connection is in TLS.

Connection can be either plaintext or TLS, and client cert can be omitted.

Jwt params for the method.

Set if mTLS is used.

List rules to select workloads that the policy should be applied on. If empty, policy will be used on all workloads in the same namespace.

List of authentication methods that can be used for peer authentication. They will be evaluated in order; the first validate one will be used to set peer identity (source.user) and other peer attributes. If none of these methods pass, request will be rejected with authentication failed error (401). Leave the list empty if peer authentication is not required

Set this flag to true to accept request (for peer authentication perspective), even when none of the peer authentication methods defined above satisfied. Typically, this is used to delay the rejection decision to next layer (e.g authorization). This flag is ignored if no authentication defined for peer (peers field is empty).

List of authentication methods that can be used for origin authentication. Similar to peers, these will be evaluated in order; the first validate one will be used to set origin identity and attributes (i.e request.auth.user, request.auth.issuer etc). If none of these methods pass, request will be rejected with authentication failed error (401). A method may be skipped, depends on its trigger rule. If all of these methods are skipped, origin authentication will be ignored, as if it is not defined. Leave the list empty if origin authentication is not required.

Set this flag to true to accept request (for origin authentication perspective), even when none of the origin authentication methods defined above satisfied. Typically, this is used to delay the rejection decision to next layer (e.g authorization). This flag is ignored if no authentication defined for origin (origins field is empty).

Define whether peer or origin identity should be use for principal. Default value is USE_PEER. If peer (or origin) identity is not available, either because of peer/origin authentication is not defined, or failed, principal will be left unset. In other words, binding rule does not affect the decision to accept or reject request.

Valid port number

Port name

Principal will be set to the identity from peer authentication.

Principal will be set to the identity from origin authentication.

exact string match.

prefix-based match.

suffix-based match.

ECMAscript style regex-based match as defined by EDCA-262. Example: “^/pets/(.*?)?”

The name must be a short name from the service registry. The fully qualified domain name will be resolved in a platform specific manner.

Specifies the ports. Note that this is the port(s) exposed by the service, not workload instance ports. For example, if a service is defined as below, then 8000 should be used, not 9000.

kind: Service
metadata:
  …
spec:
  ports:
  - name: http
    port: 8000
    targetPort: 9000
  selector:
    app: backend

Leave empty to match all ports that are exposed.