CIF

[Former Member]

Hi All,

We have created a process chain to Create the Model and to Activate the Model using Program "RIMODGEN " . In This program we have created variants for different location wise.This process chain is having 13 processess(For Each locaiton) which is running in parallel. Some time we are getting the error messege " Program for Creating a model is locked by User <User Name>"

Subramanian.K

[tibor_nagy]

Dear Subramanian.K,

I am sure that already the deletion of obsolete and processed change pointers will improve the performance a lot. You can use bd22 for this, afterwards rerun your process chains.

The parallelization only works for the selection on R/3-side and the creation or update of Products in APO.

You may review the notes 436687 and 416475.

To avoid a overload of the system you can use the parameter settings which are described in SAP Note 384077 and 74141. Make sure that you reserve a number of dialog workprocesses for online users by setting parameter rdisp/rfc_min_wait_dia_wp.

Also, review SAP Note 563806 FAQ: APO CIF. This note contains frequently asked questions about the APO CIF topic.

I hope this help.

Regards,

Tibor

[Former Member]

Note 563806 - FAQ: APO CIF

Questions regarding the integration model:

1. We have 50 integration models for each object type, since we have 50 plants. Should we define fewer integration models?

Before PlugIn 2002.1, we recommend that you define fewer models for performance reasons. Generally, the size of the integration models depends on the data volume for each plant. To optimize the integration model number, we recommend that you commission a purchase consulting service.

As of Plug-In 2002.1, the "Runtime version of the integration model" is available. Using the runtime version guarantees better performance in the online operation (also refer to the documentation for the RCIFIMAX report).

Even though the number of integration models does not affect the performance significantly, we recommend that you keep the number of integration models low, in order not to increase the runtime for generating the runtime model.

That is, do not regularly create new integration models, rather only create new versions of integration models.

You can find release notes for the PlugIn on the SAP Service Marketplace at: "http://service.sap.com/R3-PLUG-IN" -> Media Center -> Release Notes PI 2002.1 Release Notes SAP APO.

2. Do we have to transfer the master data of the vendor together with the stock data, so that consignment stocks are transferred?

Yes - this ensures that the consignment stock is correctly linked to the vendor location in SAP APO.

3. Master record objects that were changed since the last transfer are transferred again to SAP APO by initial transfer.

Does this mean that the RCPTRAN4 report (evaluate and send change recordings) does not have to run?

And what about the RBDCPCLR report (delete change pointers) for reorganizing the change pointers?

You do not have to execute the RCPTRAN4 report in this case, since the dataset in SAP APO is up to date due to the initial data transfer.

You should use the RBDCPCLR report to delete "old" change pointers.

4. The master and movement data for a material 4711 is in two active integration models (A+B). Assuming that one of the two is deactivated - what happens then?

The master and movement data remain active. See also Note 533755 "Description of the delta logic or the program RIMODINI".

5a. What happens if you deactivate an integration model that has master record objects?

Planning in SAP APO is still possible, however, you can no longer transfer the transaction data to SAP R/3.

5b. What happens with the master and movement data in SAP APO after the master data was deactivated?

The master data remains in SAP APO.

5c. What happens with the transaction data if there is another activation?

The transaction data is transferred again. Provided that you reschedule (for example plan automatically (not for plan/manufacturing orders)), the old transaction data is deleted. Note that the integration model for the master data must also be active, if the transaction data is transferred again.

6. How do I change from small to large integration models?

You activate the large model (all data already selected in active models is not transferred again) and then deactivate the small models.

7. Why are my orders not transferred from SAP R/3 to SAP APO?

Refer to the information contained in Note 424927 "No order transfer from R/3 to APO" and check your settings accordingly.

8. My material removals are not transferred in the APO order, but the stocks change.

Refer to the information contained in Note 421940 "No reduction of order reservations in APO" and check your settings accordingly.

9. Can data be transferred from SAP to SAP APO using BTE change pointers for example from the MBEW table using user exits?

Since the APO standard system does not require data from the MBEW table, this is not transferred to the CIF during the transfer of data changes using BTE. Via BTE, data is transferred by default from SAP R/3 to SAP APO from the MARA table (plant-independent material data), the MARC table (plant-dependent material data), the MARM table (conversion of units of measure) and the MAKT table (material texts). This data is also only available in the customer exit in SAP R/3. An alternative here is the transfer of the material master changes using ALE change pointers.

For example: Copying the "floating average price/periodic unit price" (MBEV-VERPR) using the CIFMAT01 user exit does not work. For this, the BD52 Customizing must be changed and the data must be transferred using the ALE method.

10. How can I avoid overlaps and thereby inconsistencies during the integration model transfer?

If you use parallel processing for the initial data transfer, transaction data may be transferred to SAP APO before the corresponding master data is available in SAP APO. For example, you can then create in-house production orders in SAP APO without PPM even though this should not be the case. Unfortunately, this cannot be prevented technically. The integration models must be cut accordingly and scheduled in background jobs so that this does not happen. Background jobs also check whether queues have been processed correctly and without errors.

11. Where can I find information about parallel processing during the initial data transfer?

You can find release notes for the PlugIn on the SAP Service Marketplace at: "http://service.sap.com/R3-PLUG-IN" -> Media Center -> Release Notes => PI 2002.1 Release Notes SAP APO.

Application log

1. Is there a way of analyzing errors in the partner system directly from the application log?

For information about this, see the following notes:

• Note 396838 "R/3: Displaying application log from queue entry"

• Note 396839 "APO: Jump to application log from incorrect queue entry"

• Note 457399 "Branching to the application log with inbound queues"

• Note 457418 "APO: Branching to the application log with inbound queues"

2. How can I find CIF logs?

In the R/3 and APO SAP systems, you can analyze the application log using the following transactions:

SAP R/3 Transaction CFG1 (see also Note 544011) and SAP APO Transaction /N/SAPAPO/C3 (see also Note 544389).

Interactive user

Question: When do I have to create a dialog user if no ATP check is to be used?

Answer: This is necessary for analyzing the data transfer and for debugging. See also Note 352844.

As of PlugIn 2002.2, it is possible to work with separate authorizations for every application.

SNP PPMs

Question: Are SNP PPMs taken into account in the engineering change management?

Answer: No (version PlugIn 2001.2).

Questions on release statuses

1. You want to use a new SAP APO 3.1 with the same system name as your old SAP APO 3.0, which is deactivated. Does this work?

Yes, as long as the "old" APO System is deactivated. The name for a logical system (LOGSYS) can only be assigned once.

You must also consider the following: In the SAP R/3 system, unique GUIDs are created for mapping purposes between SAP R/3 and SAP APO documents. See the "CIF*MAP"R/3 tables. This may cause discrepancies during the assignment of GUIDs and documents in SAP APO when you start a new initial data transfer.

2. Does SAP APO 3.1 work with PI 2001.1?

PI 2001.2 is the minimum requirement in this case. For further questions on the PlugIn release, go to the SAP Service Marketplace. Here you will find further information at "http://service.sap.com/R3-PLUG-IN" -> Integration of SAP R/3 and mySAP.com Components.

1 SAP R/3 with several SAP APOs

Question: A client of an SAP R/3 system is to be operated with several SAP APO Systems (Release 3.0 and 3.1). Does this cause problems?

Answer: In theory, this does not cause problems. However, note the following: A planned order or production order (for example order 4711), and a PREQ (PReq 4712, pos 0010) or a sales order item can only be sent to a SAP APO system, in other words a PReq created in SAP APO system 1 is not copied to SAP APO system 2. The SAP APO systems must plan different material/plant combinations. There may be no stock transfers between the SAP APO systems.

No stock transfers should occur between the SAP APO systems.

This would cause problems because a transaction date that was sent from the R/3 system to both APO systems may transfer different updates in the retransfer from both APO systems. Even if the updates from both APO systems are the same, these cannot be processed in such a way that a consistent status is achieved afterwards.

For other objects, such as TPVS and production campaigns problems may occur because updates from the APO systems can no longer occur in an indivisible LUW. This may happen if some of the referencing transaction data comes from one APO system and some from the other.

Other problems may occur if the APO systems have different release versions. The APO release may be transfer-relevant in the R/3 outbound. In this case, it cannot be guaranteed that all target systems will always be handled in a loop for all object types before each APO release query.

qRFC monitor (Transaction SMQ1/2)

1. Can I restrict the access of the 'delete' function in Transaction SMQ1 via authorizations (the display and processing functions should still be available to the user)?

There are three authorization groups for Transactions SMQ1, SMQ2:

• * Group 1 cannot call SMQ1 SMQ2 at all.

• * Group 2 can call SMQ1 SMQ2 but it can only display it (not delete it!) and activate queues. The transaction authorization for SMQ1, SMQ2 is required for this.

• * Group 3 can call SMQ1, SMQ2 and use all functions. The value NADM must be defined for this in the S_ADMI_FCD object.

2. Is there a better display of the queues than the qRFC monitor for outbound queues (SMQ1) or inbound queues (SMQ2)?

• Yes, in SAP APO you have the SCM Queue Manager in Transaction /N/SAPAPO/CQ (see also Note 419178).

• As of SCM 4.1, you can also the CIF cockpit (Transaction /SAPAPO/CC) that provides an overview of and access to all CIF-relevant transactions and Customizing settings of the APO system and all connected ERP systems.

The CIF Cockpit

As of SCM 4.1, you can also the CIF cockpit (Transaction /SAPAPO/CC). It provides an overview of and access to all CIF-relevant transactions and Customizing settings of the APO system and all connected ERP systems.

CIF Queue Names

For a list of all current CIF queue names that are used to transfer data between ERP systems and SAP APO, refer to Note 786446.

[Former Member]

Hi,

Please try using these processes in a sequence instead of making it parallel. Because running this process in parallel will create locks between two processes.

I hope this would resolve the issue.

Regards,

Tushar Bhalla.