Contact Us 1-800-596-4880

Message Logging Policy

Policy name

Message Logging

Summary

Logs custom messages using information from incoming requests, responses from the backend, or information from other policies applied to the same API endpoint

Category

Troubleshooting

First Flex Gateway version available

v1.0.0

Returned Status Codes

No return codes exist for this policy

Summary

The Message Logging policy logs custom messages using information from incoming requests, responses from the backend, or information from other policies applied to the same API endpoint.

Prerequisites

You must have the Anypoint Platform Organization Administrators role or have permission to create or manage APIs.

Configuring Policy Parameters

Flex Gateway Local Mode

In Local Mode, you apply the Message Logging policy to your API via declarative configuration files. Refer to the following policy definition and table of parameters:

- policyRef:
    name: message-logging-flex
  config:
    loggingConfiguration: <array>
      - itemName: <string> // REQUIRED
        itemData: //REQUIRED
          message: <dataweave> // REQUIRED
          conditional: <dataweave> // OPTIONAL
          category: <string> // OPTIONAL
          level: <string> // REQUIRED
          firstSection: <bool> // OPTIONAL
          secondSection: <bool> // OPTIONAL
Parameter Required or Optional Default Value Description

loggingConfiguration

Required

N/A

Array of configurations

loggingConfiguration.itemName

Required

N/A

Configuration name

loggingConfiguration.itemData.message

Required

N/A

DataWeave expression that resolves to the message log

loggingConfiguration.itemData.conditional

Optional

true

DataWeave expression that resolves to whether the message should be logged

loggingConfiguration.itemData.category

Optional

Default category

Log message prefix that indicates the package and class where the log is executed

loggingConfiguration.itemData.level

Required

N/A

Defines the level of message to log: INFO, WARN, ERROR, or DEBUG
For more information, see Message Logging Severity Levels.

loggingConfiguration.itemData.firstSection

Optional

false

Boolean that indicates if messages are logged before the API endpoint is called, taking into account the order in which this policy is applied

loggingConfiguration.itemData.secondSection

Optional

false

Boolean that indicates if messages are logged after the API endpoint is called, taking into account the order in which this policy is applied

Resource Configuration Example

- policyRef:
    name: message-logging-flex
  config:
    loggingConfiguration:
      - itemName: "Config 1"
        itemData:
          message: "#[payload]"
          conditional: "#[attributes.queryParams['QueryParam']=='someValue']"
          level: "ERROR"
          firstSection: true
      - itemName: "Config 2"
        itemData:
          message: "#['literal string']"
          level: "WARN"
          secondSection: true
          firstSection: true

Flex Gateway Connected Mode

When you use the UI to apply the Message Logging policy to your API, you can configure the following parameters:

Message Logging Policy UI
Parameter Description Required?

Message

A DataWeave expression to use for extracting the message to be logged

Required

Conditional

A DataWeave Boolean expression used to specify whether a message is to be logged

Optional

Category

A prefix in the log message that indicates the package and class where the log is executed.

Optional

Level

The level of message to log: INFO, WARN, ERROR, or DEBUG.
For more information, see Message Logging Severity Levels.

Required

Before Calling API

Set to cause your logging policy to log messages before the API endpoint is called, taking into account the order in which this policy is applied

Optional

After Calling API

Set to cause your logging policy to log messages after the API endpoint is called, taking into account the order in which this policy is applied

Optional

Example of Message Logging Policy Configured from the UI

To better understand how to use the Message Logging policy, consider a fictional retail clothier named Les Vetments. To successfully run its e-commerce portal, Les Vetments must audit the status code of the requests made on its website.

For example, if a customer buys or returns a product, or if the company adds or removes a product from its catalog, the transactions must be logged. The IT manager at Les Vetments applies the Message Logging policy to the company APIs to enable logging and viewing any data or transaction changes without making any modifications to the code.

To log the attributes of the incoming and outgoing HTTP message, Les Vetments applies the Message Logging policy to the API with the following UI configurations:

  • Message: #[attributes]

    The attributes in the payload must be logged.

  • Conditional: Blank

    Because Les Vetments requires all attributes, no expression to filter messages is specified.

  • Category: Blank

    No prefix is required in the log sentence.

  • Level: Info

    All types of messages are to be logged. The Info log level also includes Warn, Error, or Debug log messages.

  • Before Calling API: Checked

    All transactions that occur before the API is called are to be logged.

  • After calling API: Checked

    All transactions that occur after the API is called are to be logged.

Following this configuration, whenever customers access the Les Vetments catalogue or add or remove items from their carts, or when an item is added or removed from the Les Vetments catalog, logs are generated:

**********************************************************************
* Policy: message-logging-1351146-proxy                              *
* OS encoding: UTF-8, Mule encoding: UTF-8                           *
*                                                                    *
**********************************************************************
21:56:50.147     11/30/2020     Worker-0     [MuleRuntime].uber.06: [message-logging-771181-proxy].771181-message-logging.CPU_LITE @71625864     INFO
event:184152a0-3370-11eb-b732-0a8c1820c088 org.mule.extension.http.api.HttpRequestAttributes
{
   Request path=/proxy/1
   Raw request path=/proxy/1
   Method=GET
   Listener path=/proxy/*
   Local Address=/172.25.159.101:8081
   Query String=
   Relative Path=/proxy/1
   Masked Request Path=/1
   Remote Address=/18.191.37.179:21836
   Request Uri=/proxy/1
   Raw request Uri=/proxy/1
   Scheme=http
   Version=HTTP/1.1
   Headers=[
      host=logging-policy.us-e2.cloudhub.io
      x-real-ip=204.14.236.154
      accept=*/*
      user-agent=curl/7.54.0
      x-forwarded-for=204.14.236.154
      x-forwarded-port=80
      x-forwarded-proto=http
      x-sigsci-agentresponse=200
      x-sigsci-mac=7caf3820a5c07d06ef827f1565678167
   ]
   Query Parameters=[]
   URI Parameters=[]
}
21:56:50.254     11/30/2020     Worker-0     [MuleRuntime].uber.07: [logging-policy].proxy.CPU_LITE @f0ce617     INFO
event:184152a0-3370-11eb-b732-0a8c1820c088 org.mule.extension.http.api.HttpResponseAttributes
{
   Status Code=200
   Reason Phrase=OK
   Headers=[
      date=Tue, 01 Dec 2020 00:56:50 GMT
      content-type=application/json; charset=utf-8
      set-cookie=__cfduid=d8afa23a4627ebef39cbc54fea223cb231606784210; expires=Thu, 31-Dec-20 00:56:50 GMT; path=/; domain=.typicode.com; HttpOnly; SameSite=Lax
      x-powered-by=Express
      x-ratelimit-limit=1000
      x-ratelimit-remaining=999
      x-ratelimit-reset=1606784265
      vary=Origin, Accept-Encoding
      access-control-allow-credentials=true
      cache-control=max-age=43200
      pragma=no-cache
      expires=-1
      x-content-type-options=nosniff
      etag=W/"53-hfEnumeNh6YirfjyjaujcOPPT+s"
      via=1.1 vegur
      cf-cache-status=MISS
      accept-ranges=bytes
      cf-request-id=06bd666d0a0000386b7a1fc000000001
      expect-ct=max-age=604800, report-uri="https://report-uri.cloudflare.com/cdn-cgi/beacon/expect-ct"
      report-to={"endpoints":[{"url":"https:\/\/a.nel.cloudflare.com\/report?s=qUAIXoQ7qqqvFzjcwGrmK2r2PcSfDZ75jFJd0Gi0BLBUMHAcnC9wJ9I%2FEHJtk%2Bra%2FXWqkA%2F5%2FlzoWoC6YS2Lew%2BqKgjhaphHqx5WrZ0CKHMalhcM9it%2Fks0qzQGwbsRzQg%3D%3D"}],"group":"cf-nel","max_age":604800}
      nel={"report_to":"cf-nel","max_age":604800}
      cf-ray=5fa8d9c1a8c6386b-IAD
   ]
}
12:18:31.770     11/18/2020     Worker-0     agw-policy-set-deployment.01     INFO
Applied policy message-logging-1351146 version 1.0.0 to API testLoggingNew-v1-v1:16481163 (16481163) in application messagelog

To log headers, Les Vetments uses the following code snippet in the Message field:

#[attributes.headers]

To customize the log messages based on the event and the action performed, Les Vetments uses the following code snippet in the Message field:

#['User ' authentication.clientId ' performed action ' attributes.method ' on ' attributes.requestPath ' with Payload: ' ++ payload]

How This Policy Works

Logging a policy consists of the following sequence:

  1. A request is sent to the API.

  2. The message is logged if the following conditions are true:

    • The level and category of the log defined in the policy are included in the logger defined in the configuration file.

    • Either the conditional field of the policy was not set, or was set and the condition was met.

  3. The API response is returned. Message Logging Policy does not interfere with the execution of any policy nor its flow.

  4. The message appears in the application log.

About the Logging Format

A message log uses the following format:

<date> [flex-gateway-envoy][<level>] wasm log <policy-name>.default.<api-name>.default.svc main: [req: <request uuid>] [accessLog] <message>

  • policy-name: displays the name of the policy binding:

  • api-name: displays the name of the api instance:

  • request uuid: displays a unique identifier for the request that triggered the message.

Logging States During Events

Message logging occurs at specific intervals during a request flow:

  • Before logging an API flow:

    State before logging an API flow
  • After logging an API flow:

    State after logging an API flow
  • Error handler after API flow:

    State after the error handler executes

Message Logging Severity Levels

When you configure a Message Logging policy for your API instance, you can assign a severity level. Each severity level includes a specific set of information in the log report.

Log reports include the severity level you select and the following levels on the hierarchy.

For example, if you select DEBUG, the log reports include the DEBUG severity level as well as the INFO, WARN, and ERROR severity levels. However, if you select ERROR, the log reports include only error messages. The following table describes the type of information included in the logs for each severity level:

Log Severity Level Description

DEBUG

Tracing information used by application developers.

INFO

Informational messages that highlight the progress of the application.

WARN

Potentially harmful situations that indicate potential problems.

ERROR

Events that prevent normal program execution but might allow the application to continue running.