DSXchange

It remains true that the package installer can only be used by a DataStage administrator (on UNIX that means dsadm or root, on Windows it means a member of the Administrators group). It also remains true that only released jobs can be bundled into a package and therefore that only released jobs can ...

Nothing in DataStage is outside the file system, so you can use operating system backup utilities. The uvbackup utility can also be used. All backups must be performed with no DataStage processes running, lest you get a corrupted backup image caused by a dynamic hashed file changing shape half way t...

Check that the spelling and casing of the parameter name is exactly the same as you use in DSSetParam. Make sure that you really do have a DSSetParam call for this parameter (otherwise the job will use that parameter's default value). Experiment by changing that default value.

Kim's part way there. The data pathname is the pathname of the actual data portion, not of its parent directory, similarly the dictionary pathname is the pathname of the actual dictionary portion, not its parent directory. And, yes, you do need to define a primary key, which can be either a column c...

Confirmed that input links to InfxCLI stage (DS 6.0, version 0.1.3 of the InfxCLI stage) have both array size and transaction size fields. This is on the Properties tab if you've chosen Grid Style. Output links from this stage type do not have transaction handling (but it's not really relevant to SE...

DSJE.BADVALUE usually means that a value that is inappropriate for the parameter. For example, if the parameter is an Integer, then a value of 10.0784 would be illegal. But it should be perfectly OK for a parameter of type Float. Can you check in code that DataStage really believes that this paramet...

Also be aware of the -jobstatus option, which acts like -wait but also reports various information about the job run onto stdout when the job completes.

Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

This is a perfect case where a Transform is ideal. I call this Transform NowAsTimeStamp. It has a single input argument, which it ignores. It generates a TimeStamp. Its derivation expression is: Oconv(Date(),"D-YMD[4,2,2]"):" ":Oconv(Time(),"MTS:") Any place I need it, I can re-use this component. I...

If you want to append an element to a dynamic array you can use the Replace function. This can be useful in a stage variable in which you are constructing the dynamic array. Initialize the stage variable to "" (for each group of data). The derivation expression for the stage variable is Replace(stag...

Strange as it may seem a Sequential File stage can do this for you. stuff -----> Transformer -----> SeqFile -----> stuff On the output link from the Transformer stage define one column, containing tax1 : "," : value1 : "," : anothervalue1 : LF : tax2 : "," : value2 : "," : anothervalue2 LF is a stag...

Does that mean you included my copyright notice? [8D] All joking aside, the last couple of pieces of more complex code that I have posted have included such a notice, and I would appreciate it if people using it preserved that notice. The code is posted with the intention that you can use it, but wo...

Hah! [^] If you only want one pairing resolved you don't even need to create a Routine. You could use a straightforward (!) output column derivation. Let's imagine, though, that you want to maximize flexibility by creating a Routine - a transform function - to which is passed the entire string as th...

One thing that can assist the refresh is that the clients all have various filters, which have the effect of limiting the amount of information requested from the server. Sure it doesn't change the fact that it's a fat client, but it's never made any pretence of being a thin client. There's an awful...

If I were a DataStage support analyst the first question I would ask is "what's changed?" and I would not accept "nothing" as an answer. Such a quantum change in behaviour is not generated spontaneously; further, if nothing changed in your DataStage environment then it may reasonably be argued that ...

If the Informix IDS instance is on the same machine as the DataStage server, an array size of 1 will work. Otherwise, calculate array size as (packet size / row size), or an integer multiple thereof, to get efficient use of network packets. An array size of 1 will still work, but will not make very ...