DSXchange

You could do worse than to inspect the generated code (on the Job Sequence's job properties Job Control tab) to see if there's anything there that might explain the delay you are reporting.

AuditStage is still around, but will be absorbed into Information Analyzer with effect from version 8.1 if current plans come to fruition. Since 8.1 is only months away they've probably taken it off the web site so that no over-enthusiastic sales person sells any more of it.

My guess - though let's wait for the OP to get back - is that the requirement is to compare the number in the header and the number in the trailer.

Is the fastname property set to "issaq" for all node definitions in the configuration file?

Post-process your job to detect whether it sent zero rows and, if so, run another job to write the required information.

You can do this in an after-job subroutine or in a later activity in the job sequence from which the job was run.

Talk to your DB2 DBAs about client software called DB2 Connect.

Alternatively they are part of the ODBC API published by Microsoft.

The DataStage BCI is a library of functions that mimic those in the ODBC API.

No, in general. (But you can insert Transformer stages.) Personally I don't believe in aborting jobs.

Yes, provided that the metadata change involves an implicit conversion. If the metadata change involves an explicit conversion (for example string to date) then a Modify or Transformer stage will be required.

A status of Ready means that that link has not yet begun processing any rows. Therefore it is reasonable that no row counts are displayed. Later it will move briefly to a status of Starting then to Running and, later, to some form of finished/aborted.

Since writes are buffered, even setting auto-commit will not guarantee that the delete will occur first. If writes were not buffered your theory would be OK but this is not the case.

If the configuration file does not check out there's no point attempting to run a job using that configuration file.

The error suggests that one of the fastname values (issaq?) could not be resolved as a network node name.

U is not encountering the problem; Nasimul is.

The second person personal pronoun in Engish is spelled "you".

Get ETLStats from Kim Duke's website - it will do all you require (and more) without you re-inventing the wheel.

Is there any warning of the form "non numeric where numeric required, zero used" in the job log or anywhere else? (I'm puzzling over the zero in the RT_CONFIG0 name still.)