Symptom
Extracting data from larger tables (e.g., CDPOS, BKPF, CDHDR) from SAP ERP source systems to Signavio Process Intelligence could result in Exception in Advanced Extractor occurred by com.signavio.pex.sap.exception.ConnectorException java.net.SocketTimeoutException timeout.
Cause
Extracting a large amount of data in a single extraction can result in the timeout error.
Resolution
The largest amount of data is pulled during the first extraction. To avoid overloading the system, you can configure a partition strategy. With the partition strategy, you define to extract the data in chunks. From our online help:
The optimal partition size depends on the total number of data rows and how the data is distributed over a certain period of time. For example, you select a different strategy for data that is uniformly distributed over a period of time than for data that isn't equally distributed.
In addition, consider the following:
For each partition, an extraction is performed. One the one hand, the more partitions you specify to reduce the partition size, the more extractions are run. This can reduce system performance. On the other hand, if partitions are too large, connection timeouts to source systems can prevent the extraction.
Therefore, we recommend that you analyze the amount of extraction data and its distribution. Also, check with the preview function whether the selected partition strategy is reasonable.
There are a mixture of Standard and Advanced extraction modes used for SAP data sources depending on the Template you use. For example the template for Procure-to-Pay (SAP S/4HANA), CDPOS & BKPF uses Advanced while CDHDR uses Standard extraction. Follow the respective sections of the help documentation to see how they can be set up:
In short - increase the number of partitions to avoid connection timeouts. However some analysis needs to be done to determine which strategy to adopt.
Keywords
KBA , BPI-SIG-PI-DS , Data Source (CSV, Connector) , Problem