You're viewing Apigee Edge documentation.
Go to the
Apigee X documentation. info
What
Generates a custom message in response to an error condition. Use RaiseFault to define a fault response that is returned to the requesting app when a specific condition arises.
For general information on handling faults, see Handling faults.
Samples
Return FaultResponse
In the most common usage, RaiseFault is used to return a custom fault response to the
requesting app. For example this policy will return a 404
status code with
no payload:
<RaiseFault name="404"> <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables> <FaultResponse> <Set> <StatusCode>404</StatusCode> <ReasonPhrase>The resource requested was not found</ReasonPhrase> </Set> </FaultResponse> </RaiseFault>
Return FaultResponse Payload
A more complex example involves returning a custom fault response payload, along with HTTP headers and an HTTP status code. In the following example the fault response is populated with an XML message containing the HTTP status code received by Edge from the backend service, and a header containing the type of fault that occurred:
<RaiseFault name="ExceptionHandler"> <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables> <FaultResponse> <Set> <Payload contentType="text/xml"> <root>Please contact support@company.com</root> </Payload> <StatusCode>{response.status.code}</StatusCode> <ReasonPhrase>Server error</ReasonPhrase> </Set> <Add> <Headers> <Header name="FaultHeader">{fault.name}</Header> </Headers> </Add> </FaultResponse> </RaiseFault>
For a list of all variables that are available for dynamically populating FaultResponse messages, see Variables reference
Handle service callout errors
About the RaiseFault policy
Apigee Edge enables you to perform custom exception handling using a policy of type RaiseFault. The RaiseFault policy, which is similar to the AssignMessage policy, lets you generate a custom fault response in response to an error condition.
Use the RaiseFault policy to define a fault response that is returned to the requesting app when a specific error condition arises. The fault response can consist of HTTP headers, query parameters, and a message payload. A custom fault response can be more useful to app developers and app end users than generic error messages or HTTP response codes.
When executed, the RaiseFault policy transfers control from the current flow to the Error flow, which then returns the designated fault response to the requesting client app. When the message Flow switches to the Error flow, no further policy processing occurs. All remaining processing Steps are bypassed, and the fault response is returned directly to the requesting app.
You can use RaiseFault in a ProxyEndpoint or a TargetEndpoint. Usually, you will attach a Condition to the RaiseFault policy. After RaiseFault executes, Apigee will perform normal fault processing, evaluating FaultRules, or if there are no fault rules defined, it terminates processing of the request.
Element reference
The element reference describes the elements and attributes of the RaiseFault policy.
<?xml version="1.0" encoding="UTF-8" standalone="yes"?> <RaiseFault async="false" continueOnError="false" enabled="true" name="Raise-Fault-1"> <DisplayName>RaiseFault 1</DisplayName> <FaultResponse> <AssignVariable> <Name/> <Value/> </AssignVariable> <Add> <Headers/> </Add> <Copy source="request"> <Headers/> <StatusCode/> <ReasonPhrase/> </Copy> <Remove> <Headers/> </Remove> <Set> <Headers/> <Payload/> <ReasonPhrase/> <StatusCode/> </Set> </FaultResponse> <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables> </RaiseFault>
<RaiseFault> attributes
<RaiseFault async="false" continueOnError="false" enabled="true" name="Raise-Fault-1">
The following table describes attributes that are common to all policy parent elements:
Attribute | Description | Default | Presence |
---|---|---|---|
name |
The internal name of the policy. The value of the Optionally, use the |
N/A | Required |
continueOnError |
Set to Set to |
false | Optional |
enabled |
Set to Set to |
true | Optional |
async |
This attribute is deprecated. |
false | Deprecated |
<DisplayName> element
Use in addition to the name
attribute to label the policy in the
management UI proxy editor with a different, natural-language name.
<DisplayName>Policy Display Name</DisplayName>
Default |
N/A If you omit this element, the value of the policy's |
---|---|
Presence | Optional |
Type | String |
<IgnoreUnresolvedVariables> element
(Optional) Ignores any unresolved variable error in the Flow. Valid values: true/false.
Default true
.
<FaultResponse> element
(Optional) Defines the response message returned to the requesting client. FaultResponse uses the same settings as the AssignMessage policy (not available in Apigee Edge for Private Cloud).
<FaultResponse><AssignVariable> element
Assigns a value to a destination flow variable.
If the flow variable does not exist, then AssignVariable
creates it.
For example, use the following code to set the variable named myFaultVar
in the RaiseFault policy:
<FaultResponse> <AssignVariable> <Name>myFaultVar</Name> <Value>42</Value> </AssignVariable> ... </FaultResponse>
You can then refer to that variable in message templates later in the RaiseFault policy. Also, a policy attached to a FaultRule can then access the variable. For example, the following AssignMessage policy uses the variable set in the RaiseFault to set a Header in the fault response:
<AssignMessage enabled="true" name="Assign-Message-1"> <Add> <Headers> <Header name="newvar">{myFaultVar}</Header> </Headers> </Add> <IgnoreUnresolvedVariables>false</IgnoreUnresolvedVariables> <AssignTo createNew="false" transport="http" type="response"/> </AssignMessage>
<AssignVariable>
in the RaiseFault policy uses the same syntax as the
<AssignVariable>
element in the AssignMessage policy. Note that this functionality
is not currently available in Apigee Edge for Private Cloud.
<FaultResponse><Add>/<Headers> element
Adds HTTP headers to the error message. Note that the empty header
<Add><Headers/></Add>
does not add any header. This
example copies the value of the request.user.agent flow variable into the
header.
<Add> <Headers> <Header name="user-agent">{request.user.agent}</Header> </Headers> </Add>
Default: |
N/A |
Presence: |
Optional |
Type: |
String |
<FaultResponse><Copy> element
Copies information from the message specified by the
source
attribute to the error message.
<Copy source="request"> <Headers/> <StatusCode/> <ReasonPhrase/> </Copy>
Default: |
N/A |
Presence: |
Optional |
Type: |
String |
Attributes
<Copy source="response">
Attribute | Description | Presence | Type |
---|---|---|---|
source |
Specifies the source object of the copy.
|
Optional | String |
<FaultResponse><Copy>/<Headers> element
Copies the specified HTTP header from the source to the error message. To copy all headers,
specify <Copy><Headers/></Copy>.
<Copy source='request'> <Headers> <Header name="headerName"/> </Headers> </Copy>
If there are multiple headers with the same name, use the following syntax:
<Copy source='request'> <Headers> <Header name="h1"/> <Header name="h2"/> <Header name="h3.2"/> </Headers> </Copy>
This example copies "h1", "h2", and the second value of "h3". If "h3" has only one value, then it is not copied.
Default: |
N/A |
Presence: |
Optional |
Type: |
String |
<FaultResponse><Copy>/<StatusCode> element
The HTTP status code to copy from the object specified by the source attribute to the error message.
<Copy source='response'> <StatusCode>404</StatusCode> </Copy>
Default: |
false |
Presence: |
Optional |
Type: |
String |
<FaultResponse><Copy>/<ReasonPhrase> element
The reason description to copy from the object specified by the source attribute to the error message.
<Copy source='response'> <ReasonPhrase>The resource requested was not found.</ReasonPhrase> </Copy>
Default: |
false |
Presence: |
Optional |
Type: |
String |
<FaultResponse><Remove>/<Headers> element
Removes specified HTTP headers from the error message. To remove all the headers, specify
<Remove><Headers/></Remove>
. This example removes
the user-agent
header from the message.
<Remove> <Headers> <Header name="user-agent"/> </Headers> </Remove>
If there are multiple headers with the same name, use the following syntax:
<Remove> <Headers> <Header name="h1"/> <Header name="h2"/> <Header name="h3.2"/> </Headers> </Remove>
This example removes "h1", "h2", and the second value of "h3". If "h3" has only one value, then it is not removed.
Default: |
N/A |
Presence: |
Optional |
Type: |
String |
<FaultResponse><Set> element
Sets information in the error message.
<Set> <Headers/> <Payload> </Payload> <StatusCode/> <ReasonPhrase/> </Set>
Default: |
N/A |
Presence: |
Optional |
Type: |
N/A |
<FaultResponse>/<Set>/<Headers> element
Sets or overwrites HTTP headers in the error message. Note that the empty header
<Set><Headers/></Set>
does not set any header. This example sets
the user-agent
header to the message variable specified with the
<AssignTo>
element.
<Set> <Headers> <Header name="user-agent">{request.header.user-agent}</Header> </Headers> </Set>
Default: |
N/A |
Presence: |
Optional |
Type: |
String |
<FaultResponse>/<Set>/<Payload> element
Sets the payload of the error message.
<Set> <Payload contentType="text/plain">test1234</Payload> </Set>
Set a JSON payload:
<Set> <Payload contentType="application/json"> {"name":"foo", "type":"bar"} </Payload> </Set>
In a JSON payload, you can insert variables using the variablePrefix
and
variableSuffix
attributes with delimiter characters as shown in the following
example.
<Set> <Payload contentType="application/json" variablePrefix="@" variableSuffix="#"> {"name":"foo", "type":"@variable_name#"} </Payload> </Set>
or, as of cloud release 16.08.17, you can also curly braces to insert variables:
<Set> <Payload contentType="application/json"> {"name":"foo", "type":"{variable_name}"} </Payload> </Set>
Set a mixed payload in XML:
<Set> <Payload contentType="text/xml"> <root> <e1>sunday</e1> <e2>funday</e2> <e3>{var1}</e3> </Payload> </Set>
Default: |
|
Presence: |
Optional |
Type: |
String |
Attributes
<Payload contentType="content_type" variablePrefix="char" variableSuffix="char">
Attribute | Description | Presence | Type |
---|---|---|---|
contentType |
If contentType is specified, its value is assigned to the |
Optional | String |
variablePrefix | Optionally specifies the leading delimiter on a flow variable because JSON payloads cannot use the default "{" character. | Optional | Char |
variableSuffix | Optionally specifies the trailing delimiter on a flow variable because JSON payloads cannot use the default "}" character. | Optional | Char |
<FaultResponse>/<Set>/<StatusCode> element
Sets the status code of the response.
<Set source='request'> <StatusCode>404</StatusCode> </Set>
Default: |
false |
Presence: |
Optional |
Type: |
Boolean |
<FaultResponse>/<Set>/<ReasonPhrase> element
Sets the reason phrase of the response.
<Set source='request'> <ReasonPhrase>The resource requested was not found.</ReasonPhrase> </Set>
Default: |
false |
Presence: |
Optional |
Type: |
Boolean |
<ShortFaultReason> element
Specifies to display a short fault reason in the response:
<ShortFaultReason>true|false</ShortFaultReason>
By default, the fault reason in the response of the policy is:
"fault":{"faultstring":"Raising fault. Fault name : Raise-Fault-1","detail":{"errorcode":"errorCode"}}}
To make the message more readable, you can set the <ShortFaultReason>
element
to true to shorten the faultstring
to just the policy name:
"fault":{"faultstring":"Raise-Fault-1","detail":{"errorcode":"errorCode"}}}
Valid values: true/false(default).
Default: |
false |
Presence: |
Optional |
Type: |
Boolean |
Flow variables
Flow variables enable dynamic behavior of policies and Flows at runtime, based on HTTP headers, message content, or Flow context. The following predefined Flow variables are available after a RaiseFault policy executes. For more information about Flow variables, see Variables reference.
Variable | Type | Permission | Description |
---|---|---|---|
fault.name | String | Read-Only | When the RaiseFault policy executes, this variable is always set to the string
RaiseFault . |
fault.type | String | Read-Only | Returns the fault type in the error and if not available, an empty string. |
fault.category | String | Read-Only | Returns the fault category in the error and if not available, an empty string. |
Example use of RaiseFault
The following example uses a Condition to enforce the presence of a
queryparam
with the name zipcode
on the inbound request. If
that queryparam
is not present, the flow will raise a fault via RaiseFault:
<Flow name="flow-1"> <Request> <Step> <Name>RF-Error-MissingQueryParam</Name> <Condition>request.queryparam.zipcode = null</Condition> </Step> ... </Request> ... <Condition>(proxy.pathsuffix MatchesPath "/locations") and (request.verb = "GET")</Condition> </Flow>The following illustrates what would be in RaiseFault:
<RaiseFault name='RF-Error-MissingQueryParam'> <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables> <FaultResponse> <Set> <Payload contentType='application/json'>{ "error" : { "code" : 400.02, "message" : "invalid request. Pass a zipcode queryparam." } } </Payload> <StatusCode>400</StatusCode> <ReasonPhrase>Bad Request</ReasonPhrase> </Set> </FaultResponse> </RaiseFault>
Error reference
This section describes the fault codes and error messages that are returned and fault variables that are set by Edge 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.
Fault code | HTTP status | Cause |
---|---|---|
steps.raisefault.RaiseFault |
500 | See fault string. |
Deployment errors
None.
Fault variables
These variables are set when a runtime error occurs. For more information, see What you need to know about policy errors.
Variables | Where | Example |
---|---|---|
fault.name="fault_name" |
fault_name is the name of the fault, as listed in the Runtime errors table above. The fault name is the last part of the fault code. | fault.name = "RaiseFault" |
raisefault.policy_name.failed |
policy_name is the user-specified name of the policy that threw the fault. | raisefault.RF-ThrowError.failed = true |
Example error response
{ "fault":{ "detail":{ "errorcode":"steps.raisefault.RaiseFault" }, "faultstring":"Raising fault. Fault name: [name]" } }
Schema
Each policy type is defined by an XML schema (.xsd
). For reference, policy schemas
are available on GitHub.
Related topics
See Handling faults