DSXchange

Best, of course, if the two tables are in the same database, is to perform the UNION in the SQL you use to extract rows from that database.

OK you just have to systematic now. Trace every place that these parameters are being used ("connect identifier" is deliberately vague, as a security measure) to find out which one is wrong by the time the client (= DataStage job) attempts to connect to the Oracle client software or beyond.

Funnel stage is really UNION ALL. If what you want is UNION use a Funnel stage followed by a Remove Duplicates stage (which implies a sort in there somewhere).

Setting lock mode to WAIT is exactly the same thing, in principle, as building the dependencies into your design.

Check that your Source parameter is being passed correctly (look in the "job starting" event in the job log).

There's probably not enough space available at the time on whatever directory you have specified for scratch disk. Or, in the case of a Lookup stage, not enough space to load the reference data into memory.
Please post the entire (detailed) error message so we can get a better idea.

This message is generated by DB2. Consult the Errors reference manual for DB2, or your DBA. Perhaps the executing user lacks appropriate privileges on the table, perhaps the tablespace is full. Decoding error code 7 will reveal all!

COBOL programs to dump and receive text files and FTP to move them to and from your DataStage server.

The DB2/API stage can communicate via DB2 Connect software to any DB2 on any platform.

DataStage Enterprise MVS edition has support for IMS structures.

I'm not aware of any easy way to do this with DataStage Enterprise edition.

... or just apply a MONTH.FIRST Transform to it.

Welcome. :D

Use ExecSH as an after-job subroutine to create the trigger file (for example using echo > #TriggerFile# as the command). There is a check box "only execute after-job subroutine if job successful" - check that.

The MONTH.FIRST Transform will give you the first day of the month as an internal date.

Subtract one from this to get the last day of the previous month.

Technical limits for the product are largely undocumented. You just have to bump into them and then be advised by support that the limit exists.

This has been discussed before, as a Search would have revealed. Essentially you need to determine whether that condition obtains and react accordingly - perhaps a different job which handles to zero rows in database table scenario.