Contact Us 1-800-596-4880

JWT Validation Policy

Policy name

JWT Validation

Summary

Validates a JWT

Category

Security

First Mule version available

v4.1.0

Returned Status Codes

400 - Token was not provided in the request.

401

Signature was not validated, or is invalid.

Some of the required claims are not present or their validation failed.

Failed to parse the token.

Summary

JSON Web Token (JWT) is a URL-secure method of representing claims to be transferred between two parties. The claims in a JWT are encoded as a JSON object that is used as the payload of a JSON Web Signature (JWS), or as a JSON web encryption (JWE) structure in plain text. This enables the claims to be digitally signed and integrity protected with a message authentication code (MAC). Because the token is signed, you can trust the information and its source.

The JWT Validation policy validates the signature of the token and asserts the values of the claims of all incoming requests by using a JWT with JWS format. The policy does not validate JWT that uses JWE.

Configuring Policy Parameters

Mule Gateway

When you apply the policy to your API from the UI, a list of parameters are displayed based on whether your environment includes Mule or non-Mule applications.

Configuring the Parameters for Mule Applications

For Mule applications, the following parameters are displayed based on whether your environment includes Mule applications:

Element Description Example

JWT Origin

Specifies from where in the request the JWT will be extracted: * HTTP Bearer Authentication Header * Custom Expression

If you set this field to HTTP Bearer Authentication Header, the JWT is expected as Bearer.
If you set this field to Custom Expression, provide a DataWeave expression that returns the token.

Token Expression

If you set the JWT Origin to Custom Expression, type the DataWeave expression returning the JWT here.

#[attributes.headers['jwt']]
This expression searches the JWT in the header named jwt.

JWT Signing Method

Specify the signing method expected in the incoming JWT. The policy rejects the token if the JWT has a different signing method.

RSA, HMAC, ES, None

JWT Signing Key Length

Specify the length of the key (in the case of the HMAC algorithm) or the algorithm (in the case of RSA) used for the signing method.
Ignore this field if you selected none as JWT Signing Method.

256, 384, 512. For ES method, Flex Gateway only supports 256 key length.

JWT Key Origin

Specifies where to obtain the key for the Signature validation.
You can provide the key in the policy by selecting the Text option or by obtaining it from a JWKS.
Ignore this field if you selected none as JWT Signing Method.

JWT Key

This field appears if you selected Text as JWT Key Origin.
Use this field to provide the key used to check the signature of the token.
Ignore this field if you selected none as JWT Signing Method.

Enter a 32, 48, or 64 character string shared secret for HMAC, a 32 character string shared secret for ES, or the PEM Public Key without the header and footer for RSA.

JWKS URL

This field appears if you selected the JWKS method as JWT Key Origin.
Ignore this field if you selected none as JWT Signing Method.

The URL to the JWKS server.

JWKS Caching Time To Live

The amount of time, in minutes, that the JKWS is valid. When the JKWS expires, the policy retrieves it again. The devault value is 60 minutes.
Ignore this field if you selected none as JWT Signing Method.

This field input is the amount of time, in minutes, during which the policy considers the JWKS valid.

JWKS Service connection timeout (milliseconds)

Sets the maximum time, in milliseconds, to wait for a response when authenticating the access token validation endpoint.
The default value is 10 seconds.

Skip Client ID Validation

If you check this field, the policy does not verify that the client ID extracted from the JWT matches a valid client application of the API.

Client ID Expression

If Skip Client Id Validation is not set, the client ID needs to be extracted from the token.

By default, the value will be extracted using the expression #[vars.claimSet.client_id] as specified in the Oauth 2.0 token exchange draft.

Validate Audience Claim

Indicates that the policy should check for the validity of the audience claim. You can set this "Mandatory" if you select Audience Claim Mandatory.

Validate Expiration Claim

Indicates that the policy should check for the validity of the expiration claim. You can set this claim as "Mandatory" by selecting Expiration Claim Mandatory.

Validate Not Before Claim

Indicates that the policy should check for the validity of the Not Before claim. You can set this claim as "Mandatory" by selecting Not Before Claim Mandatory

Validate Custom Claim

Enables the usage of custom validations in the policy. The JWT will be valid only if all DataWeave expressions are fulfilled.

The policy provides a claimSet variable that contains all the claims present in the incoming JWT. For example:

foo : #[vars.claimSet.foo == 'fooValue']

Note that all DataWeave expressions must return a boolean value or they will always fail.

Configuring the Parameters for Anypoint Service Mesh (Non-Mule Applications)

For Anypoint Service Mesh (non-Mule applications), you configure the JWT Validation policy in the same way, excluding the following differences:

  • The policy does not accept DataWeave expressions for claim validations.

  • The token retrieving and client ID validation parameters differ, as explained in the following table.

Element Description Example

JWT Origin

Specifies from where in the request to extract the JWT:

  • jwt Header to extract the JWT from that header

  • HTTP Bearer Authentication Header for the JWT to be expected as bearer

  • Custom Header to obtain the name of the header where the JWT is expected as input

Token Header

If you set the JWT Origin to Custom Header, specify the name of the header containing the JWT.

myJwtHeader

Client ID Claim

If Skip Client Id Validation is not set, the client ID is extracted from the token.

By default, this value is extracted from the client_id claim.

How This Policy Works

The policy validates claims and signatures differently, as explained in this section.

Validating Signatures

The policy validates the signature of the JWT based on the values specified in the policy configuration. If the algorithm specified in the policy configuration does not match with the token, or if the signature of the token is invalid, all JWTs are rejected by the policy. If no algorithm is specified, the policy matches every signed and unsigned token.

The algorithms that support signature verification are:

  • Symmetric algorithms - HMAC using SHA-256, SHA-384, and SHA-512.

  • Asymmetric algorithms - ES and RSA using SHA-256, SHA-384, and SHA-512.

  • None - No signature validation.

For more information about the algorithms in a JWS structure, refer to the JWA RFC standard.

Validating Claims

Claim validations enable you to choose the conditions under which a token received in the policy is rejected. The following registered claim validations are provided by default:

  • aud: The Audience validation specifies that a token must be rejected if it does not contain at least one of the values defined.

  • exp: The Expiration validation specifies that a token must be rejected if its date is past the validation date.

  • nbf: The Not before validation specifies that the token must be rejected if the validation time is before the time the token has.

In addition to these provided claims, you can also specify other claims to use in your validations. For all claims, Registered or Custom, you must provide the following details:

  • The name of the claim you want to validate.
    For example iss, for the issuer of the token.

  • The value used to test.
    You can provide a simple literal value if you only need to verify it, or you can provide a DataWeave expression for more complex comparisons.

You can define each claim validation as mandatory or non-mandatory. If a claim is defined as mandatory and is not present in the incoming JWT, the policy rejects this token. If a claim is defined as non-mandatory and is not present in the incoming JWT, the policy does not reject the token for that specific validation.
For both cases, if a claim is present, the policy validates the token value. If the validation fails, the JWT will be rejected.

Propagating Claims

After the policy parses and validates the JWT, the claims are propagated to downstream policies and flows through the claims property in the authentication object.

To access the claim values you can use the following Dataweave 2.0 expression by replacing <claimName> with the name of the desired claim to access:

#[authentication.properties.claims.<claimName>]

The JWT is also propagated downstream, and you can access it by using the following Dataweave 2.0 expression:

#[authentication.properties.jwt]

Retry Mechanism

If you selected JSON Web Key Set (JWKS) as JWT Key Origin, the policy uses a retry mechanism to handle failed JWKS fetches. Every time the cached JWKS expires or is absent and a new request is initiated, a request is sent to the JWKS service. If the request fails, a sequence of asynchronous requests with an exponential backoff delay is sent to the JWKS service until a new JWKS is obtained and cached again.

Until the recaching finishes, the expired JWKS is used to process the request, and a 503 error is returned if no request is found. Using an expired JWKS enables you to apply cached values in upcoming requests, even if the previous requests failed.

Obtain a Policy Compatible PEM Public Key

If you have a preexisting RSA key in an unsupported format that you want to use, first covert the key to a PEM public key. The JWT policy uses public RSA key in PEM format.

Task Prerequisite

Before you convert your existing key:

  1. Obtain the public.pem key.

  2. Before you paste the key to the policy configuration, remove the -----BEGIN PUBLIC KEY----- header and the -----END PUBLIC KEY----- footer.

The following table provides instructions for how to convert your existing key based on its current format:

Starting Point Procedure

No previous certificate

  1. Generate the private RSA key openssl genrsa -out key.pem 2048

  2. Extract the public key from the generated private key key.pem openssl rsa -in key.pem -outform PEM -pubout -out public.pem

  3. The public key will be in the file public.pem

PEM Certificate

  1. Extract the public key from the PEM certificate openssl x509 -inform pem -in cert.pem -pubkey -noout > public.pem

  2. The public key will be in the file public.pem

DER Certificate

  1. Extract the public key from the DER certificate openssl x509 -inform der -in cert.crt -pubkey -noout > public.pem

  2. The public key will be in the file public.pem

x5c field from JWKS

  1. Copy the first certificate from the chain and decode it echo <certificateString> | base64 -D > cert.crt (replace <certificateString> with the copied value)

  2. Extract the public key from the DER certificate openssl x509 -inform der -in cert.crt -pubkey -noout > public.pem

  3. The public key will be in the file public.pem