DSXchange

Source and target stage are not in the same DS_JOBOBJECTS record, as you've noticed. It might be a better approach to UNION two queries (one for source stages and one for target stages). The technically correct approach is to determine the link records in DS_JOBOBJECTS, on which the DSRIDs of the st...

Why do you want to implement High Availability?

Having both servers in the same location introduces a single point of failure (for example, what happens in the case of a lightning strike?). Do re-think your strategy.

Let's begin with the second question. The number of warnings before the job aborts is an option that can be specified in Director or at run time (for example the -warn option of the dsjob command). There is no real reason to run ISD jobs in sequential mode although, fairly clearly, it's sensible to ...

Hashed File stage can not "look up ... reading from ... SQL tables".

A Transformer stage performs a lookup from a Hashed File stage. The Hashed File may have been pre-loaded with content from the table.

You need to download the Eclipse plug-in for SubVersion.

Welcome aboard. The first thing to determine is precisely what values are being passed as parameters. Inspect the "job starting" event in the job log, and let us know what you see.

Usually the files in &COMO& include the date and time that the job started, in DataStage internal format, as part of the file name. Can you list a couple of the trace file names?

Do you try to view them using the Administrator client?

Space is an ASCII character too, you know.

Why not mark this thread as Resolved?

The only real question is, is the Preserve Type property available in version 8.1?

If not, convert all your timestamps to seconds from an arbitrary zero point, aggregate them, and convert them back.

Not really. Keep in mind what a survive rule is doing; it's specifying a condition under which a field in the currect (c) record will replace that field in the so-far-the-best (b) record. Median couldn't easily be supported, particularly when many steps are involved in the survive rule, since the me...

Curious choices. Did you try c.timestamp > b.timestamp?

Whatever character is defined by APT_STRING_PADCHAR will be used to pad CHAR strings.

The only way to prevent this from happening is to use VARCHAR data type.

It's not possible. You need to use Information Server Manager to move DataStage objects into and out of SubVersion.

Ah, we seem to have a nomenclature problem here. You mean "most frequently occurring set of values" rather than "maximum". Probably the easiest approach is to form a single string made up of all the source values, possible with delimiters. Then you could use an fork-join or stage...

Shouldn't the result be KEY1 KEY2 VALUE1 VALUE2 VALUE3 A Z 10 50 20 B Y 100 5 1 You should be able to use an Aggregator stage for this. Group by KEY1 and KEY2 (not forgetting to partition by KEY1) and calculate MAX for each of VALUE1, VALUE2 and VALUE3. To keep them as integers, assert Preserve Type...