Skip to Content

Payload Based Message Search in XI30 using Trex Engine .

Introduction: Messages in XI3.0 can be monitored and searched in different ways. The messages are currently monitored via the SAP GUI using Transaction: SXMB_MONI or via the Runtime Workbench Component from the Integration Builder Page. Here, the messages can be filtered based on the Start/End Date, Time, Message Id, Sender/Receiver (Party/Service, Interface Name and Namespace) and Error Category.

As of XI30-SP15, messages can be searched based on Payload data of the message using the Trex Engine Indexing functionality. The payload-based message search is based on TREX, an SAP NetWeaver engine used to index and search documents. TREX:

•           Contains an indexing mechanism and a search engine.

•           Specializes in processing mass data.

•           Supports various different document types (such as doc, pdf, XML, or txt) and languages

•           Can be used by both ABAP and Java applications

Objectives: The payload-based message search has the following objectives.

•           To search for persisted messages on all components of an integration landscape: Integration Server, Adapter Engines and ABAP business systems

•           To uniformly configure the indexing for Integration Engines and Adapter Engines.

•           To provide a common user interface to search for messages both within individual components and across all components.

Advantages: Listed below are some of the Advantages using Trex Engine for message search in XI30.

•           All Messages from both the Integration Server and the Adapter Engine are transferred to Trex for indexing.

•           Full Text Search on Payload can be established using Trex.

•           Faster Results for Message Search

Pre requisites:

•           TREX server, version 6.1 SP13 ( or higher

•           A Configuration tool to control indexing on the individual components

•           A runtime for indexing messages

•           A user interface to search for messages using the TREX index. In this case, the user Interface is integrated into Runtime Workbench

Configuration: In order to use the payload-based message search with TREX, in XI3.0 you need to:

•           Configure the Integration engine and Adapter Engine by setting up the communication with the server and configure the indexing part.

•           Index your messages.

•           Activate and apply the search function.

•           A user interface to search for messages using the TREX index. In this case, the user Interface is integrated into Runtime Workbench

Once the configuration part is completed, the user interface is embedded into the RWB with link called: “Message Search Using Index“ as shown in the figure below.

Check out for a how to guide on the configuration topic, releasing soon on SDN.

Message Search Using Index: Below are the steps to search a message based on the Payload data. The results of a TREX search query are displayed in a table along with the attributes of the messages found. The list looks similar to the message list of message monitoring of the Runtime Workbench. From here you can navigate to the corresponding message in message monitoring by clicking an entry in the list. From there, you can display additional details for the message and perform further actions (repeat, end, and so on).

Step 1. Log in to the RWB and click on the link “Message Search Using Index”. Select the option from Indexed Component. You can select either Integration engine or Adapter Engine option. Under the Filter Tab select the various options available for message search. For instance, I am selecting a message with Start/End Date = Today, and the Payload Text = “Prasad”. There are other additional filter options available for searching messages.

Step 2. Once the selection is complete, click on the Start button to search the message based on the Payload Text. Below is the test result for my example search. Select the radio button of the message and click on Message Monitoring button

Step 3. From here you can navigate to the message details and monitor the Logs and the content of the message.

Based on the payload text (“Prasad”, Last_Name tag in the Message) I had provided in the search filter, I can my message using the Trex Indexing Functionality very quickly.


•           Netweaver ’04 Installation Guide: Search and Classification (TREX) 6.1

•           SAP Exchange Infrastructure Help.

To report this post you need to login first.


You must be Logged on to comment or reply to a post.

  1. Anonymous
    Hi Prasad,

    Nice Blog about a feature we all have been waiting for a long time.


  2. Jochen Schneider
    Hi Prasad,

    this is a great feature we were waiting for a long time. Will it be also possible to include the archived messages in the TREX search?


    1. Prasad Illapani Post author
      Hi Jochen,

      For indexing, only persisted messages can be used, but not messages that are already archived. Anyhow, depending on the retention period for messages in the index, a message might be archived while it is still visible within the TREX search. In this case, the message data will be accessed through the archive, if the message was selected from the TREX result list.


  3. Anonymous

    can I use this feature also with Trex 7.0?
    Does Trex 7.0 work with XI 3.0 SP16?

    Thanks and Regards,

  4. Himanshu shekhar
    Hi Prasad,

    It was a very useful blog..thx a lot.

    I want to know what all information trex indexes from integration server or adapter engine where messages are persisted?
    does it store only payload from the message?
    once we retrieve the message from index based on payload and if we go to detail of the message it shows ‘message from database’it means it gets detail from database nt trex server.. what if persist duration of message in integration engine or adapter engine is over?
    we  still get the message cause persist duration of trex engine could be different of other xi components.


Leave a Reply