Step 4. Submit the Source Data File for the Archive Job
Step 4. Submit the Source Data File for the Archive Job
The FILE_PATH value in the master XML file specifies the directory where the flat file or files is located. The name pattern specifies which flat files are candidate files. Ssaservice continuously checks this directory for candidate flat files. You can either create the data files in this directory or copy the flat files into this directory.
After you create the signal file
<BCP_NAME>.ok
, ssaservice picks up the data file for processing. For example, if the naming pattern for table TAB1 is
data*.csv
, the flat file
data1.csv
is a candidate file for TAB1. You must create the signal file as
data1.csv.ok
.
If the naming pattern for table TAB1 is
test*.txt
, the flat files
test1.txt
,
test22.txt
, and
test54.txt
are candidate files for TAB1. You must create the signal files as
test1.txt.ok
,
test22.txt.ok
, and
test54.txt
.
When you schedule the first load job, the executor checks whether or not the table exists. If the table exists, the load job continues. If the table is not found in Data Vault, the uploader creates and runs the DDL before the start of the loader process.
If the table attributes do not match with the table definition in the metadata, the bulk file uploader tries to alter the table with the latest metadata information from the master XML file. The bulk file uploader only attempts to alter the table if the newer column attribute is bigger in magnitude than the older type.