DSXchange

In addition, a sparse lookup, since it is executed for each row as it is read, is also used to read the current value from the source, unlike the normal lookup which has an image of the value from the time it was loaded at the start of the job run.

One option is to use the Convert() and UpCase() functions.
Note this allows spaces as well.

This sounds too much like an interview question to warrant answering in DSXChange. Since DSXChange consists of loose a group of professionals the concensus is that we prefer that DataStage developers do not memorize interview questions and answers in order to secure a role, as that is to the general...

Stage -> Record Options -> Record Delimiter.

To add to Craig's post - execute the command "orchadmin rm {full-path-to-dataset-descriptor-file}"

I would wager (but just a small amount) that either #YIL# or #AY# are in some way different between your DataStage job and your manual SELECT.
Also, if you open the monitor window do you see the same number of records going into your transform stage as are coming out?

There is no limit in that sense in DataStage. What is your SELECT clause? Could rows be dropping from the Transform stage?

DataStage will allow you to dynamically load and execute such rules, but you will be restricted to functions/rules that can be executed within the scope of a modify stage; unfortunately the transform stage is not part of the OSH mechanism directly but gets compiled from c source and is not dynamical...

Sparse lookups work correctly and are the way to solve this problem. Try running your job in just 1 node - the sparse lookup will read the new value as soon as it is written and committed to the database.

the format is to use single quotes so that the '$' in the environment variable is not expanded, i.e.

I think you have a point about overriding the defaults in the Oracle Enterprise stage - either set the value of the 2 environment variables higher than the number of expected rows/seconds of runtime or use another stage, such as the connector stage, which will allow you to specify the commit frequen...

In the sequential file properties tab, click on 'options' and you'll see that you have a property call 'File Name Column', activate that and choose the column name that you would like to use.

I used UniVerse commands at previous sites, it takes just a couple of minutes of work but involves a bit of internal knowledge. I suggest doing the following - from UNIX, get a list of a RT_STATUSnnn and RT_CONFIGnnn file names and then create a text file in the format: RESIZE {filename} 2 1 1 RESIZ...

There are several tools available, all AIX systems have 'vmstat' available and optional tools such as 'topas' might be installed as well. It might be best to ask your system administrator for some help, since the output of the tools can be a bit cryptic upon first use.

Can you not try using a file pattern in the sequential file stage, which will also give you the file name as an additional column if you so desire.