Skip to Content

With the release of version 2.47 of SAP Cloud Platform Integration, we changed the representation and the interaction with the message processing log to improve the analysis of the message processing in case of errors. This blog gives an overview of the new feature and provides best practices and examples.

In this new version we provide new features such as:

  • Condensed log data which correlate to the integration flow model
  • Simple navigation through the log data along the message processing steps.
  • Improved visualization of log.

We also outline how the new capabilities can be used in the context of a troubleshooting use case.

 

Troubleshooting always begins with the Monitor Message Processing feature and multiple search options are available to retrieve the message you are interested in.

After having selected the specific message processing log, you see the first details of the message processing. In this case we are investigating the error that occurred in the message processing. The Error Details are giving you a first technical hint on what happened.  To further analyze this issue, click on the Log Level you are seeing in the detailed view of the message processing log. In this specific case the message was sent in log level Trace.

On the left-hand side, you see the master list containing the Run Steps. A click on one of the elements will affect the views on the right-hand side, the Integration Flow Model, the Log Content and Message Content views that are selectable by clicking on the respective tabs.

The visible log steps are the logical steps that match the configured integration message flow model steps, in this case ODataSender, Script, Groovy Script and HCIOData. Other steps are not visible, since the occurring error stopped the message processing.

If the integration flow model is still deployed it will be loaded and shown as the default page.

The Run Steps table on the left side is the starting point for all views. The  indicates the steps where an error occurred. The table is sorted by occurrence, the newest entries being at the top. Errors are most likely shown in one of the last steps and that is what we are looking for in this case.

A click on this icon opens a small popover showing additional error information.

The selection of one run step affects the data shown on all 3 views that you select by clicking on the tabs above the Integration Flow Model.

In case of a split or multicast, steps belonging to one sub-message are grouped together in a segment. The segment number indicates the sub-message this step belongs to and allows to group them together. Suppose we have multiple segments and the step shows that an error occurred on segment four, we then should first start looking at the steps of this segment instead of looking at the others.

In case the Integration Flow Model is no longer deployed, this view will not be loaded and the corresponding tab is not available. By selecting a specific step from the Run Steps list, the related integration flow model shape will be highlighted. In addition, the configuration of this step is loaded below the model, as it is in the Integration Flow Model Designer. The envelope shows the message route. A filled envelope is shown if the message was processed with log level Trace and the trace data is available. Otherwise it is hollow. The red envelope shows where the error occurred.

If you select a shape within the model, the configuration for the selected shape is shown.

We can visually follow the message path by clicking on multiple steps from the bottom to the top. The Log Content tab contains the step’s Properties and the Activities list. The activities listed are the technical message processing log steps that were visible as the Run Steps in the previous design. In the past, these steps, e.g. one entry for each setting of a header or other technical information, were visible in the Run Steps view making it harder for end-users to logically map specific steps to an integration flow model step.

The content updates based on the selected Run Step allow you to go through each step and see the changes and the errors (if any) that occurred.

The Message Content tab is only visible in case the message was processed with the log level Trace. The data visible here is the same as in the previous design of the detail page.

By being able to fluently switch between all information without having to leave the page, it is much more convenient to find the required information as well as to analyze an error.

I hope you will enjoy using the new design and changes that we showcased. In case of questions or feedback feel free to comment on this blog.

 

Further References:

Trace

Previous Table View

To report this post you need to login first.

3 Comments

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

  1. Eng Swee Yeoh

    Hi Timo

     

    Thanks for providing an update on the changes to this Web UI functionality.

     

    I provided some feedback on the previous design, and I’m glad that items 1 and 3 have been addressed by the latest design changes. However, item 2 (bi-directional navigation and search functionality) is still missing and they are sorely needed when the integration flow becomes quite big/complex with many steps.

     

    Maik Keller mentioned that bi-direction navigation between the steps and the graphical model was planned, so just wondering if that is still in the pipeline?

     

    Regards

    Eng Swee

    (0) 
    1. Maik Keller

      Thank you for your feedback.

      It is still planned to provide a feature which allows to identify the log entries associated with an item in the integration flow model. When an user selects an item in the integration flow model all the referenced log entries in the step list will be highlighted.

      In addition it is planned to provide convenience functions that allow an user to efficiently navigate through the step list. Especially for large integration flows such filter and search functions are essential.

      (0) 

Leave a Reply