Continuent Ltd

replicator.extractor.parallel-extractor.ChunkDefinitionFile

The path to a file that contains a specification of the tables that should be included in the extraction. If no file is specified then all tables and extracted for provisioning.

The format of the chunk definition file is a Comma Separated Values (CSV) file, with each line containing the schema name, optional table name, optional chunk size and optional column list to be extracted. For example:

SALES

Would extract all tables within the schema SALES:

SALES,INVOICES

Would extract only the INVOICES table from the SALES schema.

SALES,INVOICES,1000

Would extract only the INVOICES table from the SALES schema, only in chunks of 1000 rows.

To extract only specific columns, add the column list to the end of the schema, table and chunk size. For example:

SALES,INVOICES,1000,INVOICENO,CUSTOMERID,VALUE

Multiple lines can be used to define a range of schemas, tables, and chunk size combinations.

replicator.extractor.parallel-extractor.chunk_size

The chunk size defines the number of rows that are extracted from the source tables and inserted into the THL. This number should be adjusted when extracting very large rows from the source tables.

replicator.extractor.parallel-extractor.add_truncate_table

If set to true, a TRUNCATE statement is inserted into the THL before the first row data is inserted into the THL. This empties the target table before the provision data is replicated.

replicator.extractor.parallel-extractor.extract_channels

Specifies the number of simultaneous threads that will be used to extract information. Defaults to a single thread.

replicator.extractor.parallel-extractor.queue_size

The number of events buffered inside the parallel extractor before they are inserted into the THL. Defaults to 20 transactions.