IBM Support

Known Issues and Limitation for CP4D SAP connectors

Troubleshooting


Problem

This document addresses the known issues, restrictions, and limitations with CP4D SAP Connectors.
 
 

Resolving The Problem

Bulk Extract Connector: Known Issues & Limitations

Issue 1: Bulk Extract flow fail with OPEN_CURSOR_FAIL Error
 
 This error is related to an SQL query that uses the new OPEN SQL Syntax, which was introduced in version 7.40, SP05. The query cannot be executed on older systems, as this syntax is not supported in those systems. Any attempt to run a similar SQL query on such systems will result in syntax check errors.
 
Due to the increased usage of the new SQL syntax, particularly its more intricate variations, there was a need to enhance the OPEN CURSOR logic employed in the SAP CPD Bulk Extract stage, especially after version 7.40. This improvement, once implemented, will only apply to systems that are compatible with the updated syntax, namely SAP ECC6.0-EHP7 and later versions.
 
The flow executes without any issues when we use the SQL query directly, but it raises an SQL validation error when we attempt to use an SQL subquery.
 

Execution Steps:
Step 1: Design the flow with Bulk Extract connector.
Step 2: Pass connection details
Step 3: Design the flow using where condition with filter having with query inside the where clause.

Example: SELECT

MARA~BLANZ
FROM
MARA AS MARA
WHERE
MARA~BLANZ IN (SELECT MARA~BLANZ FROM MARA AS MARA)
image-20230221144905-1
 

Note: Currently, the Open Cursor logic is not supported in any release of CPD Bulk Extract Connector.

 

Issue 2:  Bulk Extract flow fails with "java/lang/OutOfMemoryError" when processing a table with huge data (~ over 1M records) and default heap size.

  • This issue occurs when the default heap size is not sufficient for processing the table having huge number of records.

Solution: Heap size option is enabled on the connector level. so now if customer faces OutOfMemory error, heap size can be increased accordingly.

Steps to increase Heap size:

  1. Login to CPD environment.
  2. Design a flow with Bulk Extract connector.
  3. Go to stage properties and scroll down to other properties.
  4. Set the Heap Size accordingly.
 

image-20230221144406-1

 

Issue 3: "Fatal Error: Warning limit 100 for the job has been reached, failing the job."

Solution:

1. Open the Flow
2. Go to Settings

image-20230217131313-1

3. Select "Run."
4. Select "No Limit."
image-20230217132435-1
 
 
Issue 4: The CPD flow is failing with the following errors: " Player 1 terminated unexpectedly" and "java.lang.OutOfMemoryError: Java heap space" while processing table with huge records (~VBAP Table) with the Auto Generate RFC Destination option selected and default heap size set to 256.
 
  • This is likely due to the default heap size of 256 is not sufficient for processing the table having huge number of records.
Solution: User is always advised to increase heap memory of connector in case he is trying to process tables which contains huge number of records.
 
 
Issue 5: Unexpected behavior has been observed during the execution of a Bulk extract Flow, where the job either passes with Zero records or results in a Timeout Error.  
 
  • The sequence of certain data types is crucial, and as a result, processing the SQL query on SAP may encounter errors when the LRAW data type is listed first in the sequence.  
Solution: By altering the sequence of both the metadata and the SQL query, this error can be successfully eliminated.
 
image 12920
 
image 12919
 

Issue 6: The job output data from the SAP Bulk Extract is rounding the final digit for fields defined with the SAP data type DF16_DEC.

Description: While running the SAP Bulk Extract job, you may observe that for DF16_DEC, the last digit of the decimal value appears to be rounded off when fetched from SAP. This behaviour identified during unit testing and has been confirmed to be handled by the SAP system itself, not by the connector. The rounding of decimal values for the DF16_DEC data type is determined by SAP’s internal data handling rules and is not influenced by the connector logic.

SAP Table Data: 

image-20250513150152-1

SAP Bulk Extract Output Data:

image-20250513150248-2

Conclusion: According to the SAP team, this behavior is expected and aligns with standard precision handling for DF16 DEC within SAP. Please consider this when validating decimal data extracted through the connector for DF16 DEC Data Type column. 

BAPI Connector: Known Issues & Limitations

Issue 1: BAPI connector flow fail with "Parameter not resolvable" Error.

Solution: The SAP BAPI connection supports secrets and vaults, but two characters — double quotes (") and backslashes (\) — are not allowed in the field name of custom secrets while creating secrets for the connection details. You cannot use these characters when setting up secrets for the SAP BAPI connection.
 

Issue 2: BAPI  connector flow does not have any activated input parameters(import/table).

Solution: For a BAPI that only has export parameters activated and table parameters activated as output (with no import parameters activated and no table parameters activated as input), you must manually create one column under the Input columns in the BAPI connector for the BAPI sequence number. This is a mandatory step to complete the BAPI connector flow when no import parameters are activated, and no table parameters are activated as input.
 
Steps to add BAPISeqNo column manually:
 
  1. Go the Input tab of the BAPI operator stage and click on the "Edit" icon under the "Columns".image-20241116150151-2
  2. Add the "BAPISeqNo" column from the details of the below snapshot, click on "Apply and return" to save the changes and review the changes in the stage.image-20241116150213-3image-20241116150242-4
 
 

Delta Extract Connector: Known Issues & Limitations

Issue 1:  Delta job aborting with error when executed in Delta Mode on EHP7 System with the error JCO_ERROR_SYSTEM_FAILURE: Invalid Entry I_TSN_CONFIRMED = {2022-...
 
 
image-20230215171609-1
 

 
Solution:  The reported error can be resolved by applying the SNote 2763247 - Invalid entry I_TSN_CONFIRMED = {...} when performing ODP delta load.
 
Issue 2: Delta BW Mode jobs are failing with the error "Contact system administrator."
 
Solution: SAP schedules report which executes failed packets every 15 mins by default. Change this setting to execute the report before Delta Extract job hits timeout.
Steps:
1.    Go to t-code SM59.
2.    Expand TCP/IP connections -> double-click the applicable TCP/IP connection -> choose the destination associated with the logical system ->Open it in change mode.
3.    From the Destination menu, go to Edit and click tRFC Options.
4.    Set Time betw. 2 tries [mins] to 1 and click continue.
5.    From the Destination menu, choose Save.
 

OData Connector: Known Issues & Limitations

Issue 1: The OData connector is displaying an SAP Packs related version (8.2.0.2), which is irrelevant for CPD environment

This is how the logs display version info details.

logs:
22 04/08/2022 12:25:22 INFO IIS-CONN-DAAPI-00099 <SAP_OData_2,0> Version Info: VERINFO_PRODUCTNAME=IBM InfoSphere Information Server Pack for SAP Applications
VERINFO_PRODUCTVERSION=8.2.0.2
VERINFO_BUILDNUMBER=084
VERINFO_BUILDDATE=2020-03-24_04:29
VERINFO_PATCHLABEL=RC2

Solution: 

This problem was specific to version up to 4.5.3 and has been fixed. It is no longer observed in the versions from 4.6 release. 

Issue 2: Error when selecting metadata using the Asset Browser for SAP OData V4 with Normal and Local connections in release 5.1.1 : “ERROR BrowsingService.getAssets: Unable to retrieve data: An unexpected error occurred of type Null pointer error: Cannot invoke "java.lang.String.contains(java.lang.CharSequence)" because the return value of "java.lang.Exception.getMessage()" is null”.

OData_v4_AB_Snapshots

 

Solution: 

In CPD release 5.1.1, an issue arises when creating new SAP OData V4 flows using the Asset Browser. This problem occurs specifically when you attempt to create a new flow via the Asset Browser. However, for existing or imported jobs where columns metadata is already populated, you can still run the flow. As a result, creating a new job using the Asset Browser is not functional in release 5.1.1 and has been resolved in subsequent releases.

Issue 3: While using the Asset Browser with the SAP OData V4 mTLS connection, the metadata for the OData service cannot be loaded, resulting in the error message: "Unable to retrieve data: The assets request failed: CDIC09999E: Internal error occurred: 0 data client error: (/IWBEP/CM_V$/CM_V4_COS/011) No authorization to access service group '/IWFND/CONFIG' null [HTTP/1.1 403 Forbidden]" Odata_Issue3

Solution: 

The SAP OData V4 mTLS connection has been introduced in release 5.1.1. If you are encountering issues loading metadata while using the SAP OData V4 mTLS connection, you can switch to a non mTLS SAP Odata connection to load the metadata successfully. After completing the job design, you can switch back your connection to the SAP Odata V4 mTLS connection to ensure authentication is handled through mTLS during job execution. Please refer below screenshot for reference.Odata_Issue3_solution

Issue 4: The SAP OData job fails with the error “Internal error occurred: General security error: Private key format is not valid PEM format” when mTLS local connection is used and the “Client Private key” or “Client certificate chain” is provided directly in the stage properties. Odata_Issue4

Solution: 

You can use Parameters and Parameter sets to pass the value of “Client private Key” or “Client certificate chain”.
 
Steps to add local Parameter in the job: 
 
  1. Click on the highlighted Parameters icon to add local parameters. Odata_Issue4_Step1
  2. Add Local parameters for the “Client Private key” and “Client certificate chain” to save the changes click on “Return to canvas”. Odata_Issue4_Step2
  3. Use the created local parameters for “Client Private key” and “Client certificate chain” and click on “Test Connection”. Odata_Issue4_Step3Odata_Issue4_Step4
  4. Run the job.
     

Issue 5: Unable to select metadata using the Asset Browser for SAP OData V4 with API Key connection

When using the Asset Browser with the SAP OData V4 connector and API Key authentication in Cloud Pak for Data, users are unable to retrieve and select metadata. The Asset Browser fails to load the service metadata and displays an error similar to the following:

"ERROR BrowsingService.getAssets: Unable to retrieve data: The assets request failed: CDICO9999E: Internal error occurred: O data client error: (/IWBEP/CM_V4_COS/011) No authorization to access service group '/IWFND/CONFIG' [HTTP/1.1 403 Forbidden]"

This issue occurs when selecting metadata using an SAP OData V4 connection configured with API Key authentication.

image

 

Solution:

API Key authentication is not supported natively for SAP OData V4 services in SAP S/4HANA Cloud Public Edition. Supported authentication mechanisms include OAuth 2.0, Communication Users, and Certificates (where applicable).

Because API Key authentication is not supported for direct SAP OData V4 service access, metadata retrieval and job creation using the Asset Browser with an API Key connection is not functional.

To successfully select metadata and create SAP OData V4 jobs, configure the connection using a supported authentication mechanism such as OAuth 2.0, Communication Users, or Certificates.

 

BW Extract Connector: Known Issues & Limitations

Issue 1: When using the SAP BW Extract connector, data extraction may fail or return incorrect results if the source system name provided in the stage property is not valid. This can happen if the value is too long or contains unsupported characters, as SAP may truncate or reject the name.

SAP BW

Solution:

Make sure the source system name follows SAP’s naming rule:

  • Use only letters (A-Z), numbers (0-9), and underscores (_).
  • Do not use special characters such as @, /, \, # etc.
  • Keep the name within 10 characters.
  • Do not start with a number.
  • Ensure the name is unique in your SAP system.

Following these guidelines will help avoid truncation issues and ensure the correct source system is used during data extraction.

Issue 2: When running multiple BW Extract jobs in parallel using same Process Chain (PC), Open Hub Destination (OHD) and Source System, inconsistent behavior is observed. The system does not reliably handle concurrent execution of jobs sharing the same configuration, which results in unpredictable result.

Solution:  

Avoid running multiple BW Extract jobs in parallel when they share same Process Chain (PC), Open Hub Destination (OHD) and Source System. Concurrent execution may lead errors such as RFC server conflicts, timeout failures, or SAP BW parameter update issues.

Issue 3: In the existing BW Extract job in CPD, when user select or deselect Skip Technical Keys, the column metadata in the Output tab is displayed based on the user’s selection; however, the underlying column metadata remains unchanged because metadata is not loaded instantaneously from SAP/Asset Browser. This behavior is a UI-driven action applicable only to certain technical fields.  

Solution: In CPD, Selecting/Deselecting the Skip Technical Keys option does not update metadata correctly and may lead to undefined behavior during data extraction. User can create a new BW Extract job with the desired Skip Technical Keys configuration.

Issue 4: In Cloud Pak for Data (CPD), the SAP BW Extract job fails when the Program ID is provided in lowercase in the SAP BW Extract connector.
The job fails with an error like: 
 “Source System is attached with different Program ID: XXXXXXX”

This behavior is observed starting from 5.3.0 Patch 4 and 5.3.1 Patch 5 onwards.  

Solution: Ensure that the Program ID is provided in uppercase in the SAP BW Extract connector stage and matches exactly with the Program ID in the SAP system. 
As a best practice, always use uppercase Program ID values to avoid inconsistencies and ensure successful job execution.
In release prior to 5.3.0 Patch 4 and 5.3.1 Patch 5, the SAP BW Extract connector generated a dynamic Program ID automatically during job execution. Users were not required to specify a Program ID manually. 
Starting with 5.3.0 Patch 4 and 5.3.1 Patch 5, support was introduced to allow users to specify a Program ID in the connector stage. When using this feature, the Program ID entered in the stage must be provided in uppercase and must match exactly with the Program ID configured in the SAP system. 

 

Common Issues: 

Issue 1: Upload Jar functionality is missing in Local Connection creation for SAP connectors

 SAP connectors rely on specific JCO library files such as sapjco3.jar, libsapjco.so, and several other jar files to establish connections with SAP servers.
 
Consequently, if no jar file has been previously uploaded to the cluster whilst attempting to create a connection through the "New Asset" feature, users will not have the option to select those for local connection creation. This, in turn, can lead to job failures. To utilize these essential jar files into Local connection feature, user must upload required library files by creating a new Asset Connection.
 

Issue 2: The columns are showing up as COLUMN1, COLUMN2, etc., instead of their correct names in the source and target connectors of the SAP connectors(BAPI, IDoc, Odata) job

image-20241020113058-1

 

Solution: 

This behaviour observed when the user select input/output file from source or target connector after configuring the SAP connectors(BAPI, IDoc, Odata). To resolve this issue you must ensure to select the input/output file in the source and target connectors before configuring the SAP connectors(BAPI, IDoc, Odata) by following the below steps.
 
Browsing the input/output files from sequential file stages
 
  1. Select and open the source “Sequential file” stage in the job. image-20241016145502-3
  2. Go to the “Output” tab and expand the “Properties”, under “File” parameter click on the “Browse” button to select the file in the stage. Alternatively, user can enter the name of the input file in the “File” field from saved Data Assets (make sure the input file is already uploaded in the Data Asset).image-20241016145540-4
  3. Select the file by browsing the ds-storage, after selecting the file then click on “Add” button to add the file. To save the changes in stage, click on the “Save” button.image-20241016150317-1image-20241016151017-2
  4. Please follow the same above steps to select the output file in the target “Sequential file” stage.
 
 
Issue 3: The test connection fails with error message “The connection could not be established: Initialization of repository destination DESTINATION1 failed: Syntax error in received X.509-client certificate (base64-coding)”
 
image-20241130134316-1
 
 

Solution: 

The test connection fails because the x509 Certificate (which is required for x509 SNC connections) is being used as a custom secret in the connection. Starting from IBM Cloud Pak for Data 4.8 release, to use x509 certificates in the SNC enabled connections, you must create "SSL Certificate" secret type using secrets and vaults.
 
image-20241130170747-2image-20241130170905-3
 
Issue 4: Length & Scale Parameters value not visible in UI for INT8, DATN SAP DATA Types or INTEGER, BIGINT, DATE CPD SQL Type columns.
 
Description: When working with SAP connectors, you may notice that for certain columns (CPD SQL data types like DATE, INTEGER and BIGINT) length and scale attribute's value is disabled in the user interface during schema inspection or configuration. This behavior is not unique to SAP connectors, similar behavior can be observed with other available connectors when handling DATE, INTEGER, and BIGINT CPD SQL types. This is governed by how Cloud Pak for Data (CPD) manages SQL type adjustments. 
 
Due to byte allocation and data storage characteristics, SAP data types such as INT8 and DATN are internally mapped within the CPD environment accordingly.
•    INT8 is mapped to BIGINT
•    DATN is mapped to DATE

image-20250513151332-3
 
Conclusion: 
In CPD, the CPD SQL types BIGINT, INTEGER and DATE have fixed lengths, and by design, their lengths are not shown in the UI. Therefore, the absence of a length field for INT8 and DATN is expected and aligned with the system's internal data type handling.
Users can safely proceed with configurations and job executions without concern, as this limitation does not affect data integrity or processing.

Issue 5: Leading zeros are added to the output data for the SAP data type DEC34_DEC to match the defined column length specified in the metadata.
 
Description: You may observe that for the SAP data type DEC34_DEC, the output values have leading zeros (i.e., zeros appended to the left of the actual numeric value). This is due to how DEC34_DEC is internally mapped to the DECIMAL SQL data type in Cloud Pak for Data (CPD). The DECIMAL type in CPD supports fixed precision and scale, allowing it to accommodate larger numeric values when necessary.

As a result, when the actual stored value is smaller than the defined precision, leading zeros are automatically added to maintain the fixed format. This behavior is expected and consistent with SAP's handling of DEC data types.
Additionally, this formatting is the same as seen in the legacy IBM InfoSphere DataStage (IIS), ensuring compatibility and consistency in how these data types are displayed across platforms.
 
SAP Table Data:
 image-20250513151555-4
SAP Connectors Output Date:
 image-20250513151635-5
Conclusion:
Users can safely proceed with configurations and job executions without concern, as this limitation does not affect data integrity or processing.

Issue 6: SAP connector jobs fails when the Secret vault contains a custom field named exactly “password”. This creates a naming conflict, with the system defined password field, causing connector to resolve wrong credentials and fail during job execution. 
 
Description: The job fails only when the custom field is named “password”. Rename the custom Secret Vault field to any name other than “password”.
Examples: “pswrd”, “sapPassword”, userPassword etc. 
 
Issue 7: When the language parameter is stored in Secret Vault and used in local connection parameters/parameter sets for SAP connectors (Bulk, BAPI, Delta, IDoc and BW Extract), job log messages appear partially masked with ***. 
 

Solution: The observed behavior is aligned with the current design of the credential-handling framework used in SAP connectors and is therefore considered expected behavior.

In platform connections, all parameters defined as credentials (username, password, client, and language) are handled uniformly. As a result, when credentials are configured using the Secret Vault, the language parameter is also required to follow the same secret-based handling, even though it is not sensitive in nature.

In local connections, there is flexibility to configure parameters independently. Therefore, the language parameter can be maintained as a standard (non-secret) property, which ensures that log messages are not masked.

In summary, the behavior should be understood and handled based on connection type:

  • In platform connections, the language parameter follows credential handling and may be stored as a secret.
  • In local connections, the language parameter can be excluded from secrets and treated as a normal configuration property.
 

Known Issues:

Issue: With the latest release of SAP NWRFCSDK Patch Level 18 (PL18) on 5th December 2025, SAP BAPI flows fail in both on-Cloud Pak for Data (CPD) environment due to ICU library dependencies.

Resolution: Copy the required ICU library files to the ds-px-default-ibm-datastage-px-runtime-<identifier> pod  by using this technote: https://www.ibm.com/support/pages/node/7156762. This ensures that all runtime dependencies are satisfied and allows successful execution of the flows. 

Document Location

Worldwide

[{"Type":"MASTER","Line of Business":{"code":"LOB76","label":"Data Platform"},"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSHGYS","label":"IBM Cloud Pak for Data"},"ARM Category":[{"code":"a8m50000000ClVJAA0","label":"Analyze"},{"code":"a8m0z000000GoylAAC","label":"Troubleshooting"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]

Document Information

Modified date:
23 June 2026

UID

ibm16955793

Page Feedback