VerifyJWT policy

This page applies to Apigee and Apigee hybrid.

View Apigee Edge documentation.

What

Verifies a signed JWT or decrypts and verifies an encrypted JWT received from clients or other systems. This policy also extracts the claims into context variables so that subsequent policies or conditions can examine those values to make authorization or routing decisions. See JWS and JWT policies overview for a detailed introduction.

This policy is a Standard policy and can be deployed to any environment type. For information on policy types and availability with each environment type, see Policy types.

When this policy executes, in the case of a signed JWT, Apigee verifies the signature of the JWT using the provided verification key. In the case of an encrypted JWT, Apigee decrypts the JWT using the decryption key. In either case, Apigee subsequently verifies that the JWT is valid according to the expiry and not-before times if they are present. The policy can optionally also verify the values of specific claims on the JWT, such as the subject, the issuer, the audience, or the value of additional claims.

If the JWT is verified and valid, then all of the claims contained within the JWT are extracted into context variables for use by subsequent policies or conditions, and the request is allowed to proceed. If the JWT signature cannot be verified or if the JWT is invalid because of one of the timestamps, all processing stops and an error is returned in the response.

To learn about the parts of a JWT and how they are encrypted and signed, refer to RFC7519.

How

Whether the policy verifies a signed or encrypted JWT depends on the element you use to specify the algorithm that verifies the JWT:

• If you use the <Algorithm> element, the policy verifies a signed JWT.
• If you use the <Algorithms> element, the policy verifies an encrypted JWT.

Video

Watch a short video to learn how to verify the signature on a JWT.

Verify a Signed JWT

This section explains how to verify a signed JWT. For a signed JWT, use the <Algorithm> element to specify the algorithm for signing the key.

Samples for a signed JWT

The following samples illustrate how to verify a signed JWT.

<VerifyJWT name="JWT-Verify-HS256">
    <DisplayName>JWT Verify HS256</DisplayName>
    <Algorithm>HS256</Algorithm>
    <Source>request.formparam.jwt</Source>
    <IgnoreUnresolvedVariables>false</IgnoreUnresolvedVariables>
    <SecretKey encoding="base64">
        <Value ref="private.secretkey"/>
    </SecretKey>
    <Subject>monty-pythons-flying-circus</Subject>
    <Issuer>urn://apigee-edge-JWT-policy-test</Issuer>
    <Audience>fans</Audience>
    <AdditionalClaims>
        <Claim name="show">And now for something completely different.</Claim>
    </AdditionalClaims>
</VerifyJWT>

<VerifyJWT name="JWT-Verify-RS256">
    <Algorithm>RS256</Algorithm>
    <Source>request.formparam.jwt</Source>
    <IgnoreUnresolvedVariables>false</IgnoreUnresolvedVariables>
    <PublicKey>
        <Value ref="public.publickey"/>
    </PublicKey>
    <Subject>apigee-seattle-hatrack-montage</Subject>
    <Issuer>urn://apigee-edge-JWT-policy-test</Issuer>
    <Audience>urn://c60511c0-12a2-473c-80fd-42528eb65a6a</Audience>
    <AdditionalClaims>
        <Claim name="show">And now for something completely different.</Claim>
    </AdditionalClaims>
</VerifyJWT>

{
  "typ" : "JWT",
  "alg" : "RS256"
}

{
  "sub" : "apigee-seattle-hatrack-montage",
  "iss" : "urn://apigee-edge-JWT-policy-test",
  "aud" : "urn://c60511c0-12a2-473c-80fd-42528eb65a6a",
  "show": "And now for something completely different."
}

{
  "sub" : "monty-pythons-flying-circus",
  "iss" : "urn://apigee-edge-JWT-policy-test",
  "aud" : "urn://c60511c0-12a2-473c-80fd-42528eb65a6a",
  "show": "And now for something completely different."
}

The samples above use the <Algorithm> element, so they verify a signed JWT. The <PrivateKey> element specifies the key used to sign the JWT. There are other key elements as well. Which one you use depends on the algorithm specified by the value of <Algorithm>, as described in the next section.

Setting the key elements to verify a signed JWT

The following elements specify the key used to verify a signed JWT:

• <SecretKey>
• <PublicKey>

Which element you use depends on the chosen algorithm, as shown in the following table:

<SecretKey encoding="base16|hex|base64|base64url">
  <Value ref="private.secretkey"/>
</SecretKey>

<PublicKey>
  <Value ref="rsa_public_key_or_value"/>
</PublicKey>

<PublicKey>
  <Certificate ref="signed_cert_val_ref"/>
</PublicKey>

<PublicKey>
  <JWKS ref="jwks_val_or_ref"/>
</PublicKey>

Verify an encrypted JWT

This section explains how to verify an encrypted JWT. For an encrypted JWT, use the <Algorithms> element to specify the algorithms for signing the key and content.

Sample for an encrypted JWT

The following sample shows how to verify an encrypted JWT (with <Type> set to Encrypted), in which:

• The key is encrypted with the RSA-OAEP-256 algorithm.
• The content is encrypted with the A128GCM algorithm.

<VerifyJWT name="vjwt-1">
  <Algorithms>
    <Key>RSA-OAEP-256</Key>
    <Content>A128GCM</Content>
  </Algorithms>
  <Type>Encrypted</Type> 
  <PrivateKey>
    <Value ref="private.rsa_privatekey"/>
  </PrivateKey>
  <Subject>subject@example.com</Subject>
  <Issuer>urn://apigee</Issuer>
  <AdditionalHeaders>
    <Claim name="moniker">Harvey</Claim>
  </AdditionalHeaders>
  <TimeAllowance>30s</TimeAllowance>
  <Source>input_var</Source>
</VerifyJWT>

The sample above uses the <Algorithms> element, so it verifies an encrypted JWT. The <PrivateKey> element specifies the key that will be used to decrypt the JWT. There are other key elements as well. Which one you use depends on the algorithm specified by the value of <Algorithms>, as described in the next section.

Setting the key elements to verify an encrypted JWT

The following elements specify the key used to verify an encrypted JWT:

• <PrivateKey>
• <SecretKey>
• <PasswordKey>
• <DirectKey>

Which element you use depends on the chosen key encyryption algorithm, as shown in the following table:

<PrivateKey>
  <Value ref="private.rsa_privatekey"/>
</PrivateKey>

<PrivateKey>
  <Value ref="private.ec_privatekey"/>
</PrivateKey>

<SecretKey encoding="base16|hex|base64|base64url">
  <Value ref="private.flow-variable-name-here"/>
</SecretKey>

<PasswordKey>
  <Value ref="private.password-key"/>
  <SaltLength>
  <PBKDF2Iterations>
</PasswordKey>

<DirectKey>
  <Value encoding="base16|hex|base64|base64url" ref="private.directkey"/>
</DirectKey>

For more on the key requirements, see About signature encryption algorithms.

Element reference

The policy reference describes the elements and attributes of the Verify JWT policy.

Note: Configuration will differ somewhat depending on the encryption algorithm you use. Refer to the Samples for examples that demonstrate configurations for specific use cases.

Attributes that apply to the top-level element

<VerifyJWT name="JWT" continueOnError="false" enabled="true" async="false">

The following attributes are common to all policy parent elements.

<DisplayName>

<DisplayName>Policy Display Name</DisplayName>

Use in addition to the name attribute to label the policy in the management UI proxy editor with a different, natural-language name.

<Algorithm>

<Algorithm>HS256</Algorithm>

Specifies the cryptographic algorithm used to verify the token. Use the <Algorithm> element to verify a signed JWT.

RS*/PS*/ES* algorithms employ a public/secret key pair, while HS* algorithms employ a shared secret. See also About signature encryption algorithms.

You can specify multiple values separated by commas. For example "HS256, HS512" or "RS256, PS256". However, you cannot combine HS* algorithms with any others or ES* algorithms with any others because they require a specific key type. You can combine RS* and PS* algorithms.

<Algorithms>

<Algorithms>
    <Key>key-algorithm</Key>
    <Content>content-algorithm</Content>
</Algorithm>

Use the <Algorithms> element to verify an encrypted JWT. This element specifies the cryptographic algorithm for key encryption that must have been used when the encrypted JWT was created. It also optionally specifies the algorithm for content encryption.

Child elements of <Algorithms>

The following table provides a high-level description of the child elements of <Algorithms>:

Verification will fail if:

• The algorithm asserted in the alg property in the header of the encrypted JWT is different than the key encryption algorithm specified here in the <Key> element.
• The policy specifies a <Content> element, and the algorithm asserted in the enc property in the header of the encrypted JWT is different than that specified in the <Content> element.

For example, to verify an encrypted JWT and check that the key algorithm is RSA-OAEP-256, and that the content algorithm is A128GCM:

  <Algorithms>
    <Key>RSA-OAEP-256</Key>
    <Content>A128GCM</Content>
  </Algorithms>

Conversely, to verify an encrypted JWT and check that the key algorithm is RSA-OAEP-256, and not enforce a constraint on the content algorithm:

  <Algorithms>
    <Key>RSA-OAEP-256</Key>
  </Algorithms>

Key encryption algorithms

The following table lists the available algorithms for key encryption, as well as the type of key you must specify to verify a JWT using that key encryption algorithm.

See Verify an encrypted JWT for an example in which the key encryption algorithm is RSA-OAEP-256, so you use the <PrivateKey> element.

Content encryption algorithms

The VerifyJWT policy does not require that you specify an algorithm for content encryption. If you wish to specify the algorithm used for content encryption, do so with the <Content> child of the <Algorithms> element.

Regardless of the key encryption algorithm, the following algorithms - all symmetric and AES-based - are supported for content encryption:

• A128CBC-HS256
• A192CBC-HS384
• A256CBC-HS512
• A128GCM
• A192GCM
• A256GCM

<Audience>

<Audience>audience-here</Audience>

or:

<Audience ref='variable-name-here'/>

The policy verifies that the audience claim in the JWT matches the value specified in the configuration. If there is no match, the policy throws an error. This claim identifies the recipients that the JWT is intended for. This is one of the registered claims mentioned in RFC7519.

<AdditionalClaims/Claim>

<AdditionalClaims>
    <Claim name='claim1'>explicit-value-of-claim-here</Claim>
    <Claim name='claim2' ref='variable-name-here'/>
    <Claim name='claim3' ref='variable-name-here' type='boolean'/>
</AdditionalClaims>

or:

<AdditionalClaims ref='claim_payload'/>

Validates that the JWT payload contains the specified additional claim(s) and that the asserted claim values match.

An additional claim uses a name that is not one of the standard, registered JWT claim names. The value of a additional claim can be a string, a number, a boolean, a map, or an array. A map is simply a set of name/value pairs. The value for a claim of any of these types can be specified explicitly in the policy configuration, or indirectly via a reference to a flow variable.

The <Claim> element takes these attributes:

• name - (Required) The name of the claim.
• ref - (Optional) The name of a flow variable. If present, the policy will use the value of this variable as the claim. If both a ref attribute and an explicit claim value are specified, the explicit value is the default, and is used if the referenced flow variable is unresolved.
• type - (Optional) One of: string (default), number, boolean, or map
• array - (Optional) Set to true to indicate if the value is an array of types. Default: false.

When you include the <Claim> element, the claim names are set statically when you configure the policy. Alternatively, you can pass a JSON object to specify the claim names. Because the JSON object is passed as a variable, the claim names are determined at runtime.

For example:

<AdditionalClaims ref='json_claims'/>

Where the variable json_claims contains a JSON object in the form:

{
  "sub" : "[email protected]",
  "iss" : "urn://[email protected]",
  "non-registered-claim" : {
    "This-is-a-thing" : 817,
    "https://example.com/foobar" : { "p": 42, "q": false }
  }
}

<AdditionalHeaders/Claim>

<AdditionalHeaders>
    <Claim name='claim1'>explicit-value-of-claim-here</Claim>
    <Claim name='claim2' ref='variable-name-here'/>
    <Claim name='claim3' ref='variable-name-here' type='boolean'/>
    <Claim name='claim4' ref='variable-name' type='string' array='true'/>
</AdditionalHeaders>

Validates that the JWT header contains the specified additional claim name/value pair(s) and that the asserted claim values match.

An additionl claim uses a name that is not one of the standard, registered JWT claim names. The value of an additional claim can be a string, a number, a boolean, a map, or an array. A map is simply a set of name/value pairs. The value for a claim of any of these types can be specified explicitly in the policy configuration, or indirectly via a reference to a flow variable.

The <Claim> element takes these attributes:

• name - (Required) The name of the claim.
• ref - (Optional) The name of a flow variable. If present, the policy will use the value of this variable as the claim. If both a ref attribute and an explicit claim value are specified, the explicit value is the default, and is used if the referenced flow variable is unresolved.
• type - (Optional) One of: string (default), number, boolean, or map
• array - (Optional) Set to true to indicate if the value is an array of types. Default: false.

<CustomClaims>

Note: Currently, a CustomClaims element is inserted when you add a new GenerateJWT policy through the UI. This element is not functional and is ignored. The correct element to use instead is <AdditionalClaims>. The UI will be updated to insert the correct elements at a later time.

<Id>

<Id>explicit-jti-value-here</Id>
 -or-
<Id ref='variable-name-here'/>
 -or-
<Id/>

Verifies that the JWT has the specific jti claim. When the text value and ref attribute are both empty, the policy will generate a jti containing a random UUID. The JWT ID (jti) claim is a unique identifier for the JWT. For more information on jti, refer to RFC7519.

<IgnoreCriticalHeaders>

<IgnoreCriticalHeaders>true|false</IgnoreCriticalHeaders>

Set to false if you want the policy to throw an error when any header listed in the crit header of the JWT is not listed in the <KnownHeaders> element. Set to true to cause the VerifyJWT policy to ignore the crit header.

One reason to set this element to true is if you are in a testing environment and are not yet ready to handle a failure on a missing header.

<IgnoreIssuedAt>

<IgnoreIssuedAt>true|false</IgnoreIssuedAt>

Set to false (default) if you want the policy to throw an error when a JWT contains an iat (Issued at) claim that specifies a time in the future. Set to true to cause the policy to ignore iat during verification.

<IgnoreUnresolvedVariables>

<IgnoreUnresolvedVariables>true|false</IgnoreUnresolvedVariables>

Set to false if you want the policy to throw an error when any referenced variable specified in the policy is unresolvable. Set to true to treat any unresolvable variable as an empty string (null).

<Issuer>

<VerifyJWT name='VJWT-29'>
  ...
  <!-- verify that the iss claim matches a hard-coded value -->
  <Issuer>issuer-string-here</Issuer>

  or:

  <!-- verify that the iss claim matches the value contained in a variable -->
  <Issuer ref='variable-containing-issuer'/>

  or:

  <!-- verify via a variable; fallback to a hard-coded value if the variable is empty -->
  <Issuer ref='variable-containing-issuer'>fallback-value-here</Issuer>

The policy verifies that the issuer in the JWT (the iss claim) matches the string specified in the configuration element. The iss claim is one of the registered claims mentioned in IETF RFC 7519.

<KnownHeaders>

<KnownHeaders>a,b,c</KnownHeaders>

or:

<KnownHeaders ref='variable_containing_headers'/>

The GenerateJWT policy uses the <CriticalHeaders> element to populate the crit header in a JWT. For example:

{
  "typ": "...",
  "alg" : "...",
  "crit" : [ "a", "b", "c" ],
}

The VerifyJWT policy examines crit header in the JWT, if it exists, and for each header listed it checks that the <KnownHeaders> element also lists that header. The <KnownHeaders> element can contain a superset of the items listed in crit. It is only necessary that all the headers listed in crit are listed in the <KnownHeaders> element. Any header that the policy finds in crit that is not also listed in <KnownHeaders> causes the VerifyJWT policy to fail.

You can optionally configure the VerifyJWT policy to ignore the crit header by setting the <IgnoreCriticalHeaders> element to true.

<MaxLifespan>

  <VerifyJWT name='VJWT-62'>
  ...
  <!-- hard-coded lifespan of 5 minutes -->
  <MaxLifespan>5m</MaxLifespan>

  or:

  <!-- refer to a variable -->
  <MaxLifespan ref='variable-here'/>

  or:

  <!-- attribute telling the policy to use iat rather than nbf -->
  <MaxLifespan useIssueTime='true'>1h</MaxLifespan>

  or:

  <!-- useIssueTime and ref, and hard-coded fallback value. -->
  <MaxLifespan useIssueTime='true' ref='variable-here'>1h</MaxLifespan>
  ...
  

Configures the VerifyJWT policy to check that the lifespan of a token does not exceed a specified threshold. You can specify the threshold by using a number followed by a character, denoting the number of seconds, minutes, hours, days, or weeks. The following characters are valid:

• s - seconds
• m - minutes
• h - hours
• d - days
• w - weeks

For example, you can specify one of the following values: 120s, 10m, 1h, 7d, 3w.

The policy computes the actual lifespan of the token by subtracting the not-before value (nbf) from the expiry value (exp). If either exp or nbf is missing, the policy throws a fault. If the token lifespan exceeds the specified timespan, the policy throws a fault.

You can set the optional attribute useIssueTime to true to use the iat value instead of the nbf value when computing the token lifespan.

The use of the MaxLifespan element is optional. If you use this element, you can use it only once.

<PrivateKey>

Use this element to specify the private key that can be used to verify a JWT encrypted with an asymmetric algorithm. A description of possible child elements follows.

<Password>

<PrivateKey>
  <Password ref="private.privatekey-password"/>
</PrivateKey>

A child of the <PrivateKey> element. Specifies the password the policy should use to decrypt the private key, if necessary, when verifying an encrypted JWT. Use the ref attribute to pass the password in a flow variable.

<Value>

<PrivateKey>
  <Value ref="private.variable-name-here"/>
</PrivateKey>

Child of the <PrivateKey> element. Specifies a PEM-encoded private key that the policy will use to verify an encrypted JWT. Use the ref attribute to pass the key in a flow variable.

<PublicKey>

Specifies the source for the public key used to verify a JWT signed with an asymmetric algorithm. Support algorithms include RS256/RS384/RS512, PS256/PS384/PS512, or ES256/ES384/ES512. A description of the possible child elements follows.

<Certificate>

<PublicKey>
  <Certificate ref="signed_public.cert"/>
</PublicKey>

-or-

<PublicKey>
  <Certificate>
-----BEGIN CERTIFICATE-----
cert data
-----END CERTIFICATE-----
  </Certificate>
</PublicKey>

A child of the <PublicKey> element. Specifies the signed certificate used as the source of the public key. Use the ref attribute to pass the signed certificate in a flow variable, or specify the PEM-encoded certificate directly.

<JWKS>

  <PublicKey>
    <JWKS …  > … </JWKS>
  </PublicKey>

A child of the <PublicKey> element. Specifies a JWKS as a source of public keys. This will be a list of keys following the format described in IETF RFC 7517 - JSON Web Key (JWK).

If the inbound JWT bears a key ID which is present in the JWKS, then the policy will use the correct public key to verify the JWT signature. For details about this feature, see Using a JSON Web Key Set (JWKS) to verify a JWT.

If you fetch the value from a public URL, Apigee caches the JWKS for a period of 300 seconds. When the cache expires, Apigee fetches the JWKS again.

<Value>

<PublicKey>
  <Value ref="public.publickeyorcert"/>
</PublicKey>

-or-

<PublicKey>
  <Value>
-----BEGIN PUBLIC KEY-----
MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAw2kPrRzcufvUNHvTH/WW
...YOUR PUBLIC KEY MATERIAL HERE....d1lH8MfUyRXmpmnNxJHAC2F73IyN
ZmkDb/DRW5onclGzxQITBFP3S6JXd4LNESJcTp705ec1cQ9Wp2Kl+nKrKyv1E5Xx
DQIDAQAB
-----END PUBLIC KEY-----
  </Value>
</PublicKey>

A child of the <PublicKey> element. Specifies the public key to use to verify the signature on a signed JWT. Use the ref attribute to pass the key in a flow variable, or specify the PEM-encoded key directly.

<RequiredClaims>

<VerifyJWT name='VJWT-1'>
  ...
  <!-- Directly specify the names of the claims to require -->
  <RequiredClaims>sub,iss,exp</RequiredClaims>

  -or-

  <!-- Specify the claim names indirectly, via a context variable -->
  <RequiredClaims ref='claims_to_require'/>
  ...
</VerifyJWT>

The <RequiredClaims> element is optional. It specifies a comma-separated list of names of claims that must be present in the JWT payload, when verifying a JWT. The element ensures that the presented JWT contains the required claims but does not validate the contents of the claims. If any of the listed claims are not present, the VerifyJWT policy will throw a fault at runtime.

<SecretKey>

<SecretKey encoding="base16|hex|base64|base64url" >
  <Value ref="private.your-variable-name"/>
</SecretKey>
  

The SecretKey element is optional. It specifies the secret key to use when verifying a signed JWT that uses a symmetric (HS*) algorithm or when verifying an encrypted JWT that uses a symmetric (AES) algorithm for key encryption.

Children of <SecretKey>

The following table provides a description of the child elements and attributes of <SecretKey>:

<SecretKey encoding="hex" >
  <Value ref="private.secretkey"/>
</SecretKey>

<SecretKey>
  <Value ref="private.my-secret-variable"/>
</SecretKey>

<Source>

<Source>jwt-variable</Source>

If present, specifies the flow variable in which the policy expects to find the JWT to verify.

With this element, you can configure the policy to retrieve the JWT from a form or query parameter variable or any other variable. When this element is present, the policy will not strip any Bearer prefix that may be present. If the variable does not exist or if the policy does not find a JWT in the specified variable, the policy returns an error.

By default, when no <Source> element is present, the policy retrieves the JWT by reading the variable request.header.authorization and stripping the Bearer prefix. If you pass the JWT in the Authorization header as a bearer token (with the Bearer prefix), do not specify the <Source> element in the policy configuration. For example, you would use no <Source> element in the policy configuration if you pass the JWT in the Authorization header like this:

curl -v https://api-endpoint/proxy1_basepath/api1 -H "Authorization: Bearer eyJhbGciOiJ..."

<Subject>

<VerifyJWT name='VJWT-8'>
  ...
  <!-- verify that the sub claim matches a hard-coded value -->
  <Subject>subject-string-here</Subject>

  or:

  <!-- verify that the sub claim matches the value contained in a variable -->
  <Subject ref='variable-containing-subject'/>

  or:

  <!-- verify via a variable; fallback to a hard-coded value if the variable is empty -->
  <Subject ref='variable-containing-subject'>fallback-value-here</Subject>

The policy verifies that the subject in the JWT (the sub claim) matches the string specified in the policy configuration. The sub claim is one of the registered claims described in RFC7519.

<TimeAllowance>

<VerifyJWT name='VJWT-23'>
  ...
  <!-- configure a hard-coded time allowance of 20 seconds -->
  <TimeAllowance>20s</TimeAllowance>

  or:

  <!-- refer to a variable containing the time allowance -->
  <TimeAllowance ref='variable-containing-time-allowance'/>

  or:

  <!-- refer to a variable; fallback to a hard-coded value if the variable is empty -->
  <TimeAllowance ref='variable-containing-allowance'>30s</TimeAllowance>
  

The "grace period" for times, to account for clock skew between the issuer and verifier of a JWT. This will apply to both the expiry (the exp claim) as well as the not-before-time (the nbf claim). For example, if the time allowance is configured to be 30s, then an expired JWT would be treated as still valid, for 30 seconds after the asserted expiry. The not-before-time will be evaluated similarly.

<Type>

<Type>type-string-here</Type>

Describes whether the policy verifies a signed JWT or an encrypted JWT. The <Type> element is optional. You can use it to inform readers of the configuration whether the policy generates a signed or an encrypted JWT.

• If the <Type> element is present:
  • If the value of <Type> is Signed, the policy verifies a signed JWT, and the <Algorithm> element must be present.
  • If the value of <Type> is Encrypted, the policy verifies an encrypted JWT, and the <Algorithms> element must be present.
• If the <Type> element is absent:
  • If the <Algorithm> element is present, the policy assumes <Type> is Signed.
  • If the <Algorithms> element is present, the policy assumes <Type> is Encrypted.
• If neither <Algorithm> nor <Algorithms> is present, the configuration is invalid.

Flow variables

Upon success, the Verify JWT and Decode JWT policies set context variables according to this pattern:

jwt.{policy_name}.{variable_name}

For example, if the policy name is jwt-parse-token , then the policy will store the subject specified in the JWT to the context variable named jwt.jwt-parse-token.decoded.claim.sub. (For backward compatibility, it will also be available in jwt.jwt-parse-token.claim.subject)

Error reference

This section describes the fault codes and error messages that are returned and fault variables that are set by Apigee when this policy triggers an error. This information is important to know if you are developing fault rules to handle faults. To learn more, see What you need to know about policy errors and Handling faults.

Runtime errors

These errors can occur when the policy executes.

Deployment errors

These errors can occur when you deploy a proxy containing this policy.

Fault variables

These variables are set when a runtime error occurs. For more information, see What you need to know about policy errors.

Example error response

For error handling, the best practice is to trap the errorcode part of the error response. Do not rely on the text in the faultstring, because it could change.

Example fault rule

    <FaultRules>
        <FaultRule name="JWT Policy Errors">
            <Step>
                <Name>JavaScript-1</Name>
                <Condition>(fault.name Matches "InvalidToken")</Condition>
            </Step>
            <Condition>JWT.failed=true</Condition>
        </FaultRule>
    </FaultRules>