SAP CPI

Sunday, August 16, 2026

Groovy Script - IMP Points

Friday, July 25, 2025

Raise Fault Policy | API Management

What is Raise Fault Policy..?

The RaiseFault policy allows you to create custom messages in case of error conditions. This policy returns a FaultResponse to the requesting application if it encounters an error condition.

A FaultResponse can consist of HTTP headers, query parameters, and a message payload. These elements can be populated using variables. This enables you to send customized FaultResponses that are specific to the error conditions.

During execution, the RaiseFault policy transfers the message flow to the default ErrorFlow, which in turn returns the designated FaultResponse to the requesting application.

When the message flow switches to the default ErrorFlow, no further policy processing occurs. All remaining processing steps are bypassed, and the FaultResponse is returned directly to the requesting app.

Example:

We will create a API Proxy and specify end-point URL and then try to access it from Postman by using username/password.

If login credentials are correct then we will get response from the end-point, else we will get custom error message from RaiseFault policy.

API Proxy Creation:

End Point URL: https://dummy.restapiexample.com/api/v1/employees

(replace it with your URL)

Save it and go to Policies

Keep Basic Authentication Policy and Raise Fault Policy as shown below..

For Basic Authentication Policy:

<BasicAuthentication async='true' continueOnError='false' enabled='true' xmlns='http://www.sap.com/apimgmt'>
	<Operation>Decode</Operation>
	<IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables>
	<User ref='current.username'></User>
	<Password ref='current.password'></Password>
	<Source>request.header.Authorization</Source>
</BasicAuthentication>

For Raise Fault Policy:

<RaiseFault async="true" continueOnError="false" enabled="true" xmlns="http://www.sap.com/apimgmt">
    <FaultResponse>
        <Set>
            <Headers/>
            <Payload contentType="application/json">{"status" : "Error", "messege" : "401 Invalid User or Password", "Suggestion" : "Try with correct user name / password " } </Payload>
            <StatusCode>401</StatusCode>
            <ReasonPhrase>Unauthorized</ReasonPhrase>
        </Set>
    </FaultResponse>
    <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables>
</RaiseFault>

Specify Condition String:

(current.username != "YOUR_USERNAME") OR (current.password != "YOUR_PASSWORD")

Save it and deploy it.

Now test it from Postman..with Correct login credentials

With incorrect logins

Based on your requirement, you can customize RaiseFault Policy.

That's it.

Source: SAP Community

Thanks for reading :-)

Wednesday, July 2, 2025

Solace(EventMesh Replica) Setup

In this blog, I am going to explain how to setup Solace (EventMesh Replica) environment and establish connectivity with CPI and small Pub/Sub case study.

First, setup a Solace account with your mail account.

https://console.solace.cloud/login

Upon successful logged in,

Click on Cluster Manager and create a service ..(it's a self-explanatory process).

Click on service-name..

Go to Manage, where you can see Clients, Queues..etc

Click on Queues, where you can create queues and manage it.

Click on edit to change the settings..

Click on Queue and then create a Topic if you want it.

Connection details, which should be used to connect from CPI

Now, create a small iFlow ..

Specify, connection parameters for AdvancedEventMesh connector.

Now, save and deploy the iFlow.

Upon execution, you can see the payload/messages in queue(Solace)

That's all for Publishing part.

*********

Now, build a Subscription iFlow.

Specify connection parameters as created in Publishing side...

You need to deploy the iFlow and test it.

Thanks for reading :-)

Tuesday, July 1, 2025

Value - Mapping

In this blog, I am going to explain what is value-mapping, importance and how to implement it.?

Value Mapping in SAP CPI is a critical feature that enables seamless and consistent data transformation between disparate systems in an integration landscape.

In integration scenarios, you frequently encounter scenarios where a "country code" might be "US" in one system (source), "USA" in another (target).
Similarly, "material type" could be "ROH" in SAP ERP and "Raw Material" in a manufacturing execution system.

Without a centralized mapping, data becomes inconsistent, leading to errors, incorrect reporting, and operational inefficiencies.

Various features of Value-Mapping:

Centralized Management and Governance: Value Mappings are stored as separate artifacts within integration packages, providing a structured way to manage and govern conversion rules. This centralization aids in documentation, auditing, and ensuring adherence to enterprise-wide data standards.
Improved Maintainability and Agility: Value Mappings allow for dynamic updates without necessarily changing the iFlow's core logic. Business users or functional consultants (with appropriate access) can often manage these mappings directly in the CPI tenant, empowering them to react quickly to changes without requiring full developer involvement for every minor adjustment.
Enhanced Reusability: Value Mappings are reusable artifacts. Once defined, they can be used across any message mapping within the same integration package, promoting consistency and reducing development effort. If a value changes (e.g., a new country code is introduced), you only need to update the Value Mapping artifact once, and all dependent iFlows automatically inherit the change upon deployment.
Bidirectional Mapping: Value Mapping in CPI supports bidirectional lookups. This means you can not only map from a source value to a target value but also, if needed, reverse the lookup from the target value back to its original source value, which can be useful in certain integration scenarios or for reconciliation purposes.

First, create Value-Mapping

Under Package<sample> > Artifacts >Add>Value Mapping

Then, save and deploy it.

Create a simple iflow ...

Under Message Mapping, specify source meta data and target meta data, here I used same meta data.

<?xml version="1.0" encoding="utf-8"?>

<xs:schema attributeFormDefault="unqualified" elementFormDefault="qualified" xmlns:xs="http://www.w3.org/2001/XMLSchema">

<xs:element name="emprecords">

<xs:complexType>

<xs:sequence>

<xs:element maxOccurs="unbounded" name="emprecord">

<xs:complexType>

<xs:sequence>

<xs:element name="empid" type="xs:unsignedShort" />

<xs:element name="emplname" type="xs:string" />

<xs:element name="empfname" type="xs:string" />

<xs:element name="empage" type="xs:unsignedByte" />

<xs:element name="emporigin" type="xs:string" />

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:schema>

Select conversion>value mapping and specify SourceAgency/Target Agency and it's identifiers

On Failure is essential when there is no incoming field is available.

Where you can select ...below options based on your required.

Use Key,
Use Default value
Throw exception

Now, test the iflow with below given input and you will get given output.

Input Data/Payload:

<?xml version="1.0"?>

<empfname>David</empfname>

<emporigin>INDIA</emporigin>

</emprecord>

<empfname>Ramesh</empfname>

<emporigin>Germany</emporigin>

</emprecord>

<empfname>Kiran</empfname>

<emporigin>United States</emporigin>

</emprecord>

<emporigin>India</emporigin>

</emprecord>

<emporigin>Swiss</emporigin>

</emprecord>

<empfname>Pavan</empfname>

<emporigin>Brazil</emporigin>

</emprecord>

</emprecords>

Output Data/Payload:

<?xml version="1.0" encoding="UTF-8"?>

<empfname>David</empfname>

</emprecord>

<empfname>Ramesh</empfname>

</emprecord>

<empfname>Kiran</empfname>

</emprecord>

</emprecord>

</emprecord>

<empfname>Pavan</empfname>

</emprecord>

</emprecords>

Sample Output:

Note: We can achieve same thing with Fix Values feature also. But, you need to update the iflow, whenever you add Key/Value pair or update or delete it and then deploy it. So downtime will be there. So use Value-Mapping instead.

Under Conversions> Fix Values or Value Mapping

Thanks for reading :-)

Optimizing Resource Consumption

In this blog, I am going to explain the process of inspecting resource consumption in SAP CPI by using the Inspect Resource Consumption dashboard.

By analyzing key metrics such as database connections, data store usage, and monitoring storage you can identify potential issues and then optimize performance.

Dashboard Overview:

The Inspect Resource Consumption dashboard provides a snapshot of essential metrics and it displays various tiles that represent the consumption levels of different resources in your tenant. Each tile shows the overall consumption of a resource over the past 24 hours.

Click on any tile (connections/data store) for detailed information about consumption..

Database Resources

Inspect the usage of integration resources associated with the tenant database and with the database connection pool.

Connections: you can inspect resource usage of the database connections caused by integration flows.
Data Store: you can inspect resource usage of the tenant database caused by integration flows using data store operations steps.
Transactions: you can inspect resource usage of the database transactions caused by integration flows.
Monitoring Storage: you can inspect resource usage of the monitoring database storage caused by integration flows.

System Resources

Inspect system resource usage caused by your active integration flows.

Memory: you can inspect resource usage of system memory caused by integration flows.
Temporary Storage: you can inspect the storage usage of temporary files.

Content Resources

Inspect content resource usage caused by your integration artifacts.

Content Size: you can inspect the file size that integration artifacts occupy.
Integration Flows: you can inspect the integration flows with the highest resource consumption.

Important Points:

Resources should be monitored during Load / Performance testing
And, Level 2 / support team should monitor the resources in PROD tenant regularly.

Thanks for reading :-)

SFTP to DB (Multiple line items)

In this blog, I am going to explain how to retrieve the data/payload from SFTP and send it to DB (PostgreSQL).

Step 1:

Establish connectivity with SFTP

Get the SFTP server, port, user name, password from the infra team / client team

In my example,

Location: demo.wftpserver.com

Username: demo

Password: demo

FTP Port: 21

FTPS Port: 990

SFTP Port: 2222

Schedule it as per the requirement...

Time Zone is very important, if you want to schedule on particular time.

Step 2:

Establish connectivity with DB (PostgreSQL)

Refer my blog - how to establish the connectivity and sample queries..

JDBC - blog

Sender Side:

Employee XML

<?xml version="1.0"?>

<empfname>David</empfname>

</emprecord>

<empfname>Ramesh</empfname>

</emprecord>

<empfname>Kiran</empfname>

</emprecord>

</emprecords>

Employee XSD

<?xml version="1.0" encoding="utf-8"?>

<xs:schema attributeFormDefault="unqualified" elementFormDefault="qualified" xmlns:xs="http://www.w3.org/2001/XMLSchema">

<xs:element name="emprecords">

<xs:complexType>

<xs:sequence>

<xs:element maxOccurs="unbounded" name="emprecord">

<xs:complexType>

<xs:sequence>

<xs:element name="empid" type="xs:unsignedShort" />

<xs:element name="emplname" type="xs:string" />

<xs:element name="empfname" type="xs:string" />

<xs:element name="empage" type="xs:unsignedByte" />

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:schema>

Receiver Side:

Receiver JDBC - XSD

<?xml version="1.0" encoding="utf-8"?>

<xs:schema attributeFormDefault="unqualified" elementFormDefault="qualified" xmlns:xs="http://www.w3.org/2001/XMLSchema">

<xs:element name="root">

<xs:complexType>

<xs:sequence>

<xs:element name="StatementName">

<xs:complexType>

<xs:sequence>

<xs:element name="dbTableName">

<xs:complexType>

<xs:sequence>

<xs:element name="table" type="xs:string" />

<xs:element maxOccurs="unbounded" name="access">

<xs:complexType>

<xs:sequence>

<xs:element name="empid" type="xs:unsignedByte" />

<xs:element name="emplname" type="xs:string" />

<xs:element name="empfname" type="xs:string" />

<xs:element name="empage" type="xs:unsignedByte" />

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:sequence>

<xs:attribute name="action" type="xs:string" use="required" />

</xs:complexType>

</xs:element>

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:sequence>

</xs:complexType>

</xs:element>

</xs:schema>

Message Mapping:

Upload sender side xml XSD as source payload structure and Upload target side xml XSD as per target accepted structure...

Note: DB action should be specified like INSERT, UPDATE or DELETE..

Now, deploy the iflow and test it.

Note: Link all the steps and double check the connection parameters and ensure that the source and target systems are up and running, else you may encounter connectivity related exceptions.

Thanks for reading :-)

Monday, June 30, 2025

Transaction Handling .?

In this blog, we will see what is Transaction-Handling and how to use it.?

Transaction-Handling:

It refers to the management of a series of operations so they behave as a single unit of work — where either all steps succeed or the process handles failures gracefully to maintain data integrity.

When you enable transaction-handling enabled, it hold locks on resources (e.g., database records, queue entries) for their duration. While this is necessary for consistency, effective transaction management encourages short, well-defined transactions. This minimizes the time resources are locked, reducing contention and improving overall system performance and scalability.

Options:

No Transaction: Each step is committed independently. Only suitable for very simple, non-critical scenarios where data consistency is not paramount.
Required for JDBC: This ensures that all operations interacting with CPI's internal database (e.g., Data Store operations, Write Variables used with Data Stores) are part of a single transaction. If one fails, all are rolled back.
Required for JMS: Similarly, this applies transactional behavior to operations involving JMS queues managed by CPI.

Main Iflow:

Local / Child Iflow:

Transactional processing means that the message (as defined by the steps contained in a process) is processed within one transaction.

For example, consider a process with a Data Store Write operation. If transaction handling is activated, the Data Store entry is only committed if the whole process is executed successfully. In an error case, the transaction is rolled back and the Data Store entry isn't written. If transaction handling is deactivated, the Data Store entry is committed directly when the integration flow step is executed. In an error case, the Data Store entry is nevertheless persisted (and not removed or rolled back).

Note:

It's crucial to understand that CPI's transaction handling primarily covers its internal resources (like Data Stores and JMS queues).
It does not provide a distributed transaction coordinator (like a two-phase commit) across heterogeneous external systems (e.g., atomically committing changes in an SAP ERP system and an external Salesforce system).

For such complex cross-system consistency, you often need to implement patterns like Saga patterns, compensating transactions, or ensure idempotent operations on the external systems.

In conclusion, transaction handling in SAP CPI is a critical enabler for building reliable, fault-tolerant, and data-consistent integration solutions that can gracefully handle errors and maintain the integrity of your business data across interconnected systems.

Source: community.sap.com

Thanks for reading :-)