DSXchange

Perform the delete and insert operation sequentially and not in parallel. Maintain a row level commit between the operations.

The library files of the transformer stage was not copied to the compute node from the conductor node. In order to copy the library files during runtime, you need to set the APT_COPY_TRANSFORM_OPERATOR to TRUE. Else, you may also perform a rsync. A sample command to copy the files from source server...

In the XML input stage, in the Transformation Settings, tick the box for repitition element required.

Can you please clarify which load/write method and partitioning method you are using in the enterprise stage and the connector stage?

The maximum length of a varchar field can be of 255.

Please try the below logic

Trim(to_transf.STATUS) <> 'Agent'

to restrict the incoming records.

While creating the table is the table data partitioned? Then the distribution of data among database partitions is not as desired. Also, if the database is data partitioned then a better approach is to search in a particular data partition itself. Example: Suppose a table is data partitioned based o...

Try using an exception table in a DB2 connector stage. It will automatically route the duplicate primary keys to the exception table.

Are you updating any batch control table in any database? Then you may check the connection between the datastage and the database. Also check for the file system.

Yes datastage is capable of exporting executables only without GUI.

You can use transformer stage and implement the below logic in the stage variables to produce the funcionality similar to an aggregator stage. The pre-requisite to use this stage is that the data should be sorted on the keys columns that will be used for group by clause, here it is abc, def, gkt, et...

The table should be database partioned based on the column date eg. FISCAL_MONTH. This will enable the database engine to configure the data according to the partitions. Datastage will not require any special treatment to load the data in a particular partition. However, if a table is database parti...

May be the DB tables are in load pending state then.

When you are performing a lookup on 2 columns in the same stage, then the output is true only when both the conditions are met i.e. basically AND logic is being implemented. Only when you find a match for both the columns, the result is a valid one.

1. Filter all the records according to the REC_TYPE. LINK1:(REC_TYPE = '1') CUS_TYPE,CUS_ID,REC_TYPE,DATE,DETAIL C,123,1,2011/02/28,XXXA C,100,1,2011/02/28,AAA LINK2:(REC_TYPE = '2') CUS_TYPE,CUS_ID,REC_TYPE,DATE,DETAIL C,123,2,2011/02/28,XXXB C,100,2,2011/02/28,AAA1 LINK3:(REC_TYPE = 'D') CUS_TYPE,...