DSXchange

The project and job are not given as pathnames.

There is no "missing columns action" in parallel Sequential File stage.

Check again. What I described is the only possible cause for the message "string[max=0] found".

Have you "checked the same" while the job is running?

It's really an informational message that you seem always to get with decimal_from_string() even when the conversion will always be good; use a message handler to demote it to informational.

Are you using sort or hash method in the Aggregator?

Are there fields on the input of the Aggregator that are not used? If so, get rid of them.

Is this the same question as your post about Char(1) containing four characters?

Return the date in the Routine.

When assigning a value to the job parameter, use the Routine activity's return value.

This would work just as well if you used the hashed file.

The upgrade should be straightforward. Make sure you have a backup and an export. Printer segment removed is not an error - it's an informational message resulting from someone logging in with a different login ID than the user who previously had that user number, soon after the previous user logged...

There are specific stage types for working with MQ Series queues and with XML. As suggested review the manuals for these. If you're unfamiliar with the technologies, organize to be briefed by someone who is familiar, ideally with how the technology is used at that site.

All the column values that you tried to insert are given in the logged error message. You can inspect that to determine which of them is too large.

Not sure what you mean by a batch file. Do you mean a DataStage batch (created in Director) or a BAT file (created in Windows/DOS)? What, after that, do you mean by a "DataStage shell"? Why do you need to search for it? It's in the DataStage engine directory's bin sub-directory. Can you please sugge...

@TM is used to construct a multi-column key. You definitely don't need it, if JOB_NAME is the only key column, and it may be interfering with correct operation. You pass the job name to UtilityHashLookup and request it to return the field number containing START_TS. You can establish this field numb...

It's nothing to do with Unicode. Somewhere in your design you have a VarChar data type with no precision; that is, a VarChar(0).

Fix it.

The dsjob executable is in the DataStage engine's bin subdirectory. This needs to be in your PATH, just as it does in Windows, or the full pathname of the executable must be given, just as it must in Windows. So, if you're attached to the $DSHOME directory, you can use the relative pathname bin/dsjo...