SAP Community - SAP HANA Cloud

FX Revaluation - SAP S4HANA Public Cloud -Exchange rates

2026-05-20T15:12:45.424000+02:00

Dear all,

Do you happen to know whether there are any country specific rules for running the FX Revaluation (in SAP S4HANA Public Cloud), and/or Exchange rates upload schedule. For instance, if at month end the FX revaluation can be performed using the exchange rates that were uploaded in the last day of the month, after 1 pm?

Thank you,

BR,

Isabela

Profit Center Derivation Issue in New Intercompany Solution SAP S/4 HANA Public Cloud

2026-05-25T17:10:40.684000+02:00

Hi SAP Community,

After activating the new Intercompany (IC) solution in our sandbox system, we conducted testing and observed a change in system behavior related to Profit Center derivation in RV documents.

Observations:

In the new IC setup, the Profit Center for RV documents is derived from the Expense Invoice (ER).
In the old IC setup, the Profit Center was derived from the default Profit Center.

Impact:

Due to this change:

RV documents are posting to a different Profit Center.
KR documents are posting to the default Profit Center.

As a result, RV and KR documents are not aligned on the same Profit Center, whereas our requirement is that both document types should post to the same Profit Center.

As per SAP’s suggestion, we deactivated the node “Activate PC Derivation Based on Intercompany Sales Order Item (107061)”. However, this did not produce the desired results.

All combinations of test cases have been executed, and the results are attached for your reference.

We kindly request your guidance on how to ensure that both RV and KR documents derive and post to the same Profit Center.

Thanks

Naveen Kumar

SQLMD: Cross Node Joins Analysis

2026-05-26T19:31:19.805000+02:00

When executing SQLMD for Cross Node Join analysis, the program needs an input file containing Tables and Nodes.

How to generate this file using the data from HANA DB? Wherever I check in System tables/views, this field with Node information is marked as Internal Use only and always return empty result.

So, how to generate the file with Table & Node information required to execute SQLMD?

Thanks

Unable to run Hardware Configuration Check Tool (HWCCT)

2026-05-28T07:54:00.682000+02:00

Dear SAP Team,

We are currently attempting to perform hardware validation checks for our SAP HANA system. However, we are facing issues with both the SAP Hardware Configuration Check Tool (HWCCT) and some of the required OS-level validation commands.

Details are as follows:

1. We attempted to execute the HWCCT utility using:
python hwcct.py

but received the following error:
"python: can't open file 'hwcct.py': [Errno 2] No such file or directory"

2. We also searched the server for the HWCCT utility using:
find /usr/sap -iname "hwcct"

but no HWCCT files were found on the system.

3. Additionally, while performing manual hardware validation checks, some commands are unavailable on the server. For example:
numactl --hardware

returns:
"If 'numactl' is not a typo you can use command-not-found to lookup the package that contains it"

Could you please assist us with:

1. The correct procedure to obtain/install the latest HWCCT tool for our HANA version
2. Any prerequisite packages or OS dependencies required
3. Recommended SAP-supported commands or procedures for performing hardware validatio

Difference in New Invoicis Due Amount in Cash Collection Tracker App

2026-05-28T14:36:46.047000+02:00

We have identified a difference in the “New Invoices Due” amount shown in the Cash Collection Tracker under the Account Receivable Overview app in FIORI.

We need to understand the calculation logic being used by FIORI for deriving the “New Invoices Due” amount in the Cash Collection Tracker section of the Account Receivable Overview app.

Kindly check and provide the calculation logic.

T code:MB5M -ALV Grid option enabling

2026-06-02T06:37:17.049000+02:00

Inorder to trigger the business with a report of expiring or expired materials , the Standard tcode: MB5M will be scheduled in the background, however the report format is not user friendly on downloading it in Excel.

I need to provide the business it in an ALV forma so that the report is clean.

What are the potential differences in Sales Revenue of Actual Amount and Recognized Revenue?

2026-06-05T08:43:34.190000+02:00

Dear experts,

We are planning to introduce the Project Financial Booklet – Professional Services app to our business users. To support this transition, I created a bookmark that is similar to the current public bookmark used in the Projects Baseline/EAC/Ongoing app (old version of the Project Finance Booklet).

Since the Project Financial Booklet app does not provide an Actual Amount field, I have used Recognized Revenue instead. So far, the values displayed appear to be the same as the Sales Revenue values under Actual Amount in the Projects Baseline/EAC/Ongoing app.

However, because we need to explain this change to our business users, we would like to understand whether there are any differences in the definition and/or data source between these two measures.

If Sales Revenue (Actual Amount) and Recognized Revenue can produce different values, could you please explain under what circumstances such differences may occur?

Thank you in advance for your help.

Connection failure for TGKTST on DBW system (Host: vhpdphbddb01)

2026-06-05T17:53:26.408000+02:00

A connection issue is being reported from TGKTST on the DBW system on vhpdphbddb01.

A database error occured. The error message in the database was: "internal error: Remote Source [TGKTST] is not accessible: [...xODBC][Driver Manager]Can't open lib '/opt/microsoft/msodbcsql18/lib64/libmsodbcsql-18.4.so.1.1' : file not found"

Could you please verify if the ODBC driver is correctly installed on the host?

504: GATEWAY TIMEOUT error in Cloud Foundry Environment

2026-06-09T11:19:18.740000+02:00

When calling the API in the SAP BTP Cloud Foundry (CF) environment, it connects to the backend OData service (on-premises) to fetch data from the SAP system. The target endpoint "default" parameter io.timeout.millis has been configured to 60 seconds as per the SAP documentation. However, the issue is that a "504 - Gateway Timeout" error is returned by the API before 60 seconds.
Why is the io.timeout.millis configuration not being considered or meaningful in the CF environment? Could this be related to a limitation or issue in the SAP BTP Cloud Foundry environment?

https://help.sap.com/docs/sap-api-management/sap-api-management/target-endpoint-properties

Change in description of BOMs

2026-06-12T14:08:27.608000+02:00

Hello,

We are facing the following problem: when we are changing the description of the BOM for one alternative in our productive system.

This is something that should not be like that, we need to have diffrend descriptions for every alternative of the material.

BR,

Blagovesta

Do ERP Systems Really Need Machine Learning? Exploring SAP HANA PAL in SAP BTP

2026-06-15T08:59:59.941000+02:00

Introduction

Enterprise Resource Planning (ERP) systems have traditionally been designed to manage and process business transactions efficiently. Their primary role has been to capture, store, and organize business data across functions such as Sales, Procurement, Finance, Manufacturing, and Supply Chain.

However, modern organizations expect more from their ERP systems. Businesses are no longer satisfied with simply storing historical data; they want systems that can analyze information, identify patterns, predict future outcomes, and support decision-making.

This raises an important question:

Do ERP systems really need Machine Learning?

The answer depends on the business problem being addressed. Machine Learning should not be implemented merely because the technology is available. Instead, it should be used where it can provide measurable business value.

As part of my research on SAP HANA Predictive Analysis Library (PAL) Procedures in SAP Business Technology Platform (SAP BTP), I explored how built-in Machine Learning algorithms can be applied directly within the SAP ecosystem to solve real business challenges.

This blog summarizes some of the key Machine Learning capabilities available through SAP HANA PAL and their relevance in ERP scenarios.

What is SAP HANA PAL?

The Predictive Analysis Library (PAL) is a collection of Machine Learning and statistical algorithms that run directly inside the SAP HANA database.

Rather than exporting data to external Machine Learning platforms, PAL enables organizations to perform predictive analytics where the data already resides.

Some key advantages include:

Reduced data movement
Improved performance
Better data governance
Integration with SAP applications and business processes
Real-time analytics capabilities

PAL provides algorithms for:

Time Series Forecasting
Classification
Regression
Clustering
Association Rules
Outlier Detection
Recommendation Scenarios

Let's explore how these capabilities can be used in ERP environments.

1. Time Series Forecasting

Forecasting is one of the most valuable Machine Learning applications in ERP systems.

Organizations continuously need to estimate future demand, sales, inventory levels, and procurement requirements.

Using historical business data, forecasting algorithms can predict future trends and support proactive planning.

Typical ERP Use Cases

Sales Forecasting

Predict future sales based on historical transactions and seasonal patterns.

Production Planning

Estimate manufacturing demand and optimize production schedules.

Inventory Planning

Maintain optimal stock levels while reducing overstocking and stock shortages.

Procurement Planning

Predict material requirements and improve purchasing decisions.

Business Value

Improved demand planning
Reduced inventory carrying costs
Better resource utilization
Enhanced supply chain efficiency

2. Classification

Classification algorithms predict predefined categories or outcomes based on historical data.

In ERP systems, classification can help organizations estimate probabilities and support decision-making processes.

Example: Quotation Conversion Prediction

Many companies generate thousands of sales quotations every year. However, only a portion of these quotations eventually become sales orders.

Using classification models, organizations can analyze historical quotation data and predict:

Which quotations are likely to convert
Which opportunities require additional attention
Which customers have higher purchase probability

Additional ERP Use Cases

Customer Churn Prediction

Identify customers who are likely to stop doing business with the company.

Supplier Risk Classification

Classify suppliers based on delivery performance, quality metrics, and compliance factors.

Payment Default Prediction

Identify customers with a higher likelihood of delayed or missed payments.

Business Value

Improved sales effectiveness
Better risk management
Enhanced customer retention
Stronger supplier governance

3. Regression

While classification predicts categories, regression predicts continuous numerical values.

Regression algorithms are highly useful when organizations need estimates for future business metrics.

ERP Use Cases

Revenue Prediction

Forecast future revenue based on historical trends and business drivers.

Sales Value Forecasting

Estimate the expected value of upcoming sales opportunities.

Procurement Cost Estimation

Predict material procurement costs using historical pricing data.

Product Price Forecasting

Estimate future product pricing trends.

Customer Lifetime Value Analysis

Predict the long-term value of customer relationships.

Business Value

Improved budgeting and planning
Better financial forecasting
Data-driven pricing strategies
Enhanced profitability analysis

4. Association Rules

Association Rule Mining identifies relationships between products, services, and customer transactions.

This technique is widely used in retail, e-commerce, and sales analytics.

Example

Suppose customers frequently purchase:

Product A
Product B

together within the same transaction.

Association Rule algorithms can discover this pattern and provide actionable recommendations.

ERP Use Cases

Product Recommendations

Suggest related products during sales processes.

Cross-Selling Opportunities

Identify complementary products that increase sales revenue.

Product Bundling

Create optimized product packages based on purchasing behavior.

Customer Buying Pattern Analysis

Understand customer purchasing habits and preferences.

Business Value

Increased sales revenue
Improved customer experience
Better marketing effectiveness
Stronger product strategy

5. Clustering

Clustering is an unsupervised Machine Learning technique used to group similar records without predefined labels.

It helps organizations uncover hidden patterns within business data.

ERP Use Cases

Customer Segmentation

Group customers based on purchasing behavior, revenue, geography, or demographics.

Supplier Segmentation

Categorize suppliers according to performance, risk, or strategic importance.

Product Segmentation

Identify product groups with similar sales or profitability characteristics.

Regional Demand Analysis

Analyze demand patterns across different regions.

Business Value

Better customer targeting
Improved supplier management
More effective marketing campaigns
Enhanced strategic planning

Do ERP Systems Really Need Machine Learning?

The answer is:

Not every ERP process requires Machine Learning.

Core ERP functionality such as transaction processing, accounting, order management, and inventory tracking can continue operating effectively without predictive models.

However, when organizations seek answers to questions such as:

What will happen next?
Which customers are likely to buy?
Which suppliers pose risks?
How much inventory will be needed?
What revenue can be expected?

Machine Learning becomes highly valuable.

ERP systems contain vast amounts of historical business data. Machine Learning enables organizations to transform that data into actionable insights and intelligent decision support.

Key Takeaway

Machine Learning should not be implemented simply because it is available.

The real objective should always be:

"Which business problem are we trying to solve?"

When applied to the right use cases, Machine Learning can transform ERP systems from transaction-processing platforms into intelligent business platforms.

One of the most exciting aspects of SAP HANA PAL is that these capabilities can be executed directly within the SAP ecosystem, reducing the need for external Machine Learning platforms while keeping analytics close to the data.

As I continue exploring SAP HANA PAL Procedures in SAP BTP, it is becoming increasingly clear that Forecasting, Classification, Regression, Association Rules, and Clustering have significant potential to bring intelligence into ERP processes and help organizations make smarter decisions.

References

SAP HANA Predictive Analysis Library (PAL)
SAP HANA Machine Learning Capabilities
SAP Business Technology Platform (SAP BTP)
SAP HANA Cloud Predictive Analytics

Error when trying to create a service entry sheet (SES) via i_serviceentrysheettp_2

2026-06-15T22:13:21.393000+02:00

Hello,

I am encountering an error when trying to create a new service entry sheet (SES) with a position and two cost centers in one position (account assignment).
In this case, there are 3 entities involved (header, item, and account assignment).

When I have more than 1 account assignment position, I am always getting the following error:
411 06 Sum of quantities >3,000< larger than total quantity

Although my quantities are okay - 2 in item and 1 + 1 in account assignment positions.

Code example:

METHOD if_oo_adt_classrun~main.

DATA:
ls_header_data TYPE i_serviceentrysheettp_2,
ls_item_data TYPE i_serviceentrysheetitemtp_2,
ls_acc_ass TYPE I_SesAccountAssignmentTP_2,

lt_header_crt TYPE TABLE FOR CREATE i_serviceentrysheettp_2\\serviceentrysheet,
lt_item_cba TYPE TABLE FOR CREATE i_serviceentrysheettp_2\\serviceentrysheet\_serviceentrysheetitem,
lt_acc_ass_cba TYPE TABLE FOR CREATE i_serviceentrysheettp_2\\ServiceEntrySheetItem\_AccountAssignment,
ls_acc_ass_cba LIKE LINE OF lt_acc_ass_cba,
lv_item_cid TYPE string,
lv_accass_cid TYPE string.

m_out = out.

ls_header_data-serviceentrysheetname = 'Demo SES'.
ls_header_data-purchaseorder = '4500001414'.
ls_header_data-sesoriginobjecttype = `EX`.
ls_header_data-PurgDocExternalReference = 'FA163E5F512FEFE199B52BA24E8069BE'.
ls_header_data-PostingDate = '20260527'.

APPEND INITIAL LINE TO lt_header_crt ASSIGNING FIELD-SYMBOL(<ls_hdr_crt>).
<ls_hdr_crt> = CORRESPONDING #( ls_header_data CHANGING CONTROL ).
<ls_hdr_crt>-%cid = `HEADER_1` .

APPEND INITIAL LINE TO lt_item_cba ASSIGNING FIELD-SYMBOL(<ls_itm_cba>).
<ls_itm_cba>-%cid_ref = 'HEADER_1'.

lv_item_cid = |Item_2|.
lv_accass_cid = |Accass_2|.

CLEAR: ls_item_data, ls_acc_ass_cba.
ls_item_data-purchaseorderitem = '00020'.
ls_item_data-confirmedquantity = '2.000'.
ls_item_data-accountassignmentcategory = 'K'.
ls_item_data-multipleacctassgmtdistribution = '1'.
ls_item_data-ServicePerformanceDate = '20260610'.
ls_item_data-ServicePerformanceEndDate = '20260610'.

APPEND INITIAL LINE TO <ls_itm_cba>-%target ASSIGNING FIELD-SYMBOL(<item_data>).
<item_data> = CORRESPONDING #( ls_item_data CHANGING CONTROL ).
<item_data>-%cid = lv_item_cid. " 'Item_1'.

ls_acc_ass_cba-%cid_ref = lv_item_cid.
DATA(lv_accass_cid_2) = 'ACCASS_2_1'.

CLEAR ls_acc_ass.
ls_acc_ass-GLAccount = '0021351301'.
ls_acc_ass-CostCenter = '2000B11020'.
ls_acc_ass-MultipleAcctAssgmtDistrPercent = '50'.
ls_acc_ass-Quantity = '1.000'.

APPEND INITIAL LINE TO ls_acc_ass_cba-%target ASSIGNING FIELD-SYMBOL(<accass_data>).
<accass_data> = CORRESPONDING #( ls_acc_ass CHANGING CONTROL ).
<accass_data>-%cid = lv_accass_cid_2.

lv_accass_cid_2 = 'ACCASS_2_2'.

CLEAR ls_acc_ass.
ls_acc_ass-GLAccount = '0021351301'.
ls_acc_ass-CostCenter = '2000B16110'.
ls_acc_ass-MultipleAcctAssgmtDistrPercent = '50'.
ls_acc_ass-Quantity = '1.000'.

APPEND INITIAL LINE TO ls_acc_ass_cba-%target ASSIGNING <accass_data>.
<accass_data> = CORRESPONDING #( ls_acc_ass CHANGING CONTROL ).
<accass_data>-%cid = lv_accass_cid_2. " 'Accass_1'.

APPEND ls_acc_ass_cba TO lt_acc_ass_cba.

MODIFY ENTITIES OF i_serviceentrysheettp_2
ENTITY serviceentrysheet
CREATE
FROM lt_header_crt
CREATE BY \_serviceentrysheetitem
FROM lt_item_cba
ENTITY ServiceEntrySheetItem
"CREATE
" FROM lt_item_cba
CREATE BY \_AccountAssignment
FROM lt_acc_ass_cba
FAILED DATA(ls_failed_crt)
REPORTED DATA(ls_reported_crt)
MAPPED DATA(ls_mapped_crt).

COMMIT ENTITIES
RESPONSE OF i_serviceentrysheettp_2
FAILED DATA(ls_failed_commit)
REPORTED DATA(ls_reported_commit).

Maybe someone will see an error in my code and provide me a hint. Thank you.

New company codes (Field status variant issue)

2026-06-16T11:28:47.281000+02:00

Dear SAP experts,

I would like to ask you for help. We are create new company codes in out S/4 Hana Public cloud environment and we are receiving the following error message when we try to transport it from the customizing environment to the Test environment.

Unresolved dependancy to key 100 YB12 of table T004F.

I was checking the T004F in the On-premise version and it refers to the Field status variant so my first idea was whether the field status variant hadn't been assigned to the new company codes yet. But when I tried to create a new G/L account in the customizing environment, it offers me the field groups (incl. YB12) for selection.

Would you have idea what might be causing this error?

Thank you
Pavel

Feasibility Inquiry for Temporary Point‑in‑Time Restoration

2026-06-16T22:41:18.754000+02:00

I would like to request detailed information about the technical feasibility of performing a temporary point‑in‑time restoration in order to recover a table that was accidentally deleted. This request is strictly for feasibility assessment purposes, and I am not authorizing any action, restoration, or system intervention at this stage.

Additionally, I would like to understand the estimated time required to perform such a temporary point‑in‑time restoration, as well as the approximate cost associated with this type of operation, should it be technically feasible.

For clarity, I understand that a temporary point‑in‑time restoration involves restoring the full database to a previous state solely for data extraction purposes, and that the restored instance is discarded afterwards. My intention is only to confirm whether this procedure is possible in our environment and what its implications would be.

I Built a RAG Pipeline on SAP HANA Cloud Vector Engine (and Benchmarked It Against BM25)

2026-06-19T09:49:10.007000+02:00

Hey everyone,

I'm a student focused on cloud architecture. Wanted to see how SAP handles AI workloads in practice, so I built a RAG (Retrieval-Augmented Generation) pipeline from scratch on SAP HANA Cloud's Vector Engine, paired with LangChain and Google Gemini 3.5 Flash, and ran some basic accuracy benchmarks on top of it.

What I built

Standard RAG setup: load PDFs, chunk them, embed with Gemini, store the vectors in a REAL_VECTOR column in HANA Cloud. At query time, run a cosine similarity search and feed the retrieved context into Gemini for the final answer.

PDF → PyPDFLoader → RecursiveCharacterTextSplitter → Gemini Embedding → SAP HANA Cloud (REAL_VECTOR)

Connection is a plain direct hdbcli connection, nothing fancy:

conn = dbapi.connect(
address=os.getenv("HANA_DB_ADDRESS"),
port=int(os.getenv("HANA_DB_PORT")),
user=os.getenv("HANA_DB_USER"),
password=os.getenv("HANA_DB_PASSWORD"),
encrypt=True
)

One thing that tripped me up: HANA Cloud Trial blocks external connections by default, so you need to allowlist your client IP from HANA Cloud Central in the BTP Cockpit first. Worth checking before you start debugging your connection string.

Benchmarking

Since the pipeline was already there, I compared three approaches on a self-made 20-question test set: keyword search (BM25), vector search alone, and vector search + RAG.

Metric BM25 Vector Only Vector + RAG

Hit Rate@3 (doc)	55.0%	70.0%	70.0%
MRR@3	51.7%	67.5%	67.5%
Avg latency	14ms	609ms	9,511ms
Cost per query	¥0	¥0	~¥0.037

Semantic search clearly beats keyword search on accuracy (+15pt Hit Rate), but once you add the LLM generation step, latency jumps to ~9.5 seconds per query. That's a real concern for anything latency-sensitive — async processing or caching would help here.

What's next

Planning to dig into the SAP Cloud Application Programming Model (CAP) and its BTP integration next.

Full code: https://github.com/paraccoli/SAP-HANA-Cloud-Vector-Engine-RAG
Full writeup: https://portfolio.paraccoli.net/blog/sap-hana-cloud-vector-engine-rag

Curious if anyone's run into the same IP allowlist issue on Trial, or has thoughts on cutting that RAG latency down.

Scope Item Activation 5HP - Advanced Intercompany Stock Transfer

2026-07-07T14:39:59.338000+02:00

Hi SAPSupport,

could you please assist us in activating the Scope Item 5HP? We are not able to see it in our CBC customizing setup.

My colleague told me that SAP support is needed to activate this scope item.

Link: https://community.sap.com/t5/enterprise-resource-planning-blog-posts-by-sap/advanced-intercompany-stock-transfer-5hp-in-sap-s4hana-cloud-public-edition/ba-p/14178380

Thanks and BR

Dion

SAP HANA Cloud Central Gen AI is not executing most SQL queries correctly

2026-07-07T15:33:27.548000+02:00

The Gen AI-assisted database administration feature in SAP HANA Cloud Central is not executing most SQL queries correctly. The majority of queries submitted via the Gen AI virtual assistant are returning errors. This appears to be related to the feature being under active development, as noted in SAP documentation. [1] Only a limited subset of operations — specifically creating an instance and migrating a server — are functioning as expected. All other SQL generation and execution attempts are failing, significantly limiting the usability of the Gen AI feature for day-to-day database administration tasks. [2]

Bank requires SEPA format v3.7 or higher

2026-07-08T05:58:53.737000+02:00

We have recently configured the DMEE tree using the SEPA_CT format, specifically based on Z_SEPA_CT. However, our bank has informed us that they require the SEPA format version 3.7 or higher for our transactions.

Could you please provide guidance on how to update our SEPA payment format accordingly? It would be greatly appreciated if you can share the necessary steps or any new templates that align with the required version.

Thank you for your assistance.

Bank Reply-

we would like to inform you today about the planned deactivation of older SEPA format versions.

In the summer of last year, we announced in the letter "Important Changes in SEPA Payments with Commerzbank" that files in SEPA formats below Version 3.0 would no longer be accepted as of March 31, 2026.

To ensure the smoothest possible transition, a new timeline and a phased deactivation approach have been developed. The updated timeline takes into account the still high number of submissions in older formats.

Current timeline for the phased deactivation of SEPA legacy formats:

As of May 1, 2026: SCT V2.5 (Fi

The batch management requirement cannot be changed. Stocks already exist at plant l

2026-07-15T13:29:52.204000+02:00

Subject: Unable to Activate Batch Management for Existing Material

We are unable to activate Batch Management for material 31.07.0119.

Current stock is 0.
However, the system displays the message:
"The batch management requirement cannot be changed. Stocks already exist at plant level."
We verified that there is no current stock, but the material has historical goods movements.

Create KG Graph UI option is not available at SAP HANA Cloud

2026-07-16T11:25:56.015000+02:00

Hi Team,

Even after enabling triple store option at SAP HANA Cloud. I am unable see the create Knowledge graph option at SAP HANA Cloud central. Our screenshot is 348. But unable to look as image(1)

Kindly help us in providing the details about this issue.