DSXchange

My mistake, I hadn't updated the target field. Once again, the problem lays between the seat and the keyboard.

I exported the jobs and compared using Notepad++ and there was nothing there.

I checked to find if I could compare them in Datastage but the option seems to have disappeared.

What difference should I see in the DSX

Our team has noticed that all inserts to the tables which contain accented characters are failing to be mapped properly. We now need to go through hundreds of jobs and change the string fields to Unicode fields in the output of the transformer before the insert. The project default is set to UTF-8 I...

There is no warning or error given.

I tried to nest the functions but this didn't work.

I will increase the size and accept the fact that although the new field is now nullable, there is no risk of it actually being null.

It does not warrant another modify

Trying to convert a Timestamp(Microseconds) to string gives the following error modify(0),0: Un-handled conversion error on field "DT_DEB_SITUATION " from source type "timestamp[microseconds]" to destination type "string[20]": source value="2013-01-01 00:00:00.0000...

This issue was resolved. It was actually an error in a sort stage which would not allow a timestamp with microseconds as a sort key.

Converted it to a string in transformer and then it worked.

I am trying to create a generic job to remove duplicates based on a key column and a timestamp column. The key column can potentially be any numeric type, so I am creating another column in a string format and importing the key column as well as retaining the original key column The modify statement...

The updates are functioning with the custom SQL I am sorry Ray, I don't quite understand what you mean by that. Do you mean like this ? UPDATE TR_MODALITE SET DT_FIN_IMAGE = \'2015-05-18 17:48:03.000\', NUM_CHARGEMENT_UPD=20150506160002 WHERE NOM_MODALITE=? AND CODE=? AND TOP_LAST_IMAGE=1 I tried th...

I did think of that but the number of lines stored in the array are unknown. Do you have any idea how much data can be stored in the userstatus variable ?

I have a job which reads a table into a routine. The routine contains a dynamic array in a common area which stores each column individually for each row. The common uses the following syntax COMMON /x/ a b This job is called from a job sequence. Once the job has run, I have another routine to read ...

Problem solved. The table script missed out the all important .DBO

I have created an generic job to load a series of files into their counterpart tables on SQL Server. All of the tables load, except one. I get this error Odbc_Select_Schema: ODBC function "SQLExecDirect()" reported: SQLSTATE = 42S02: Native Error Code = 208: Msg = [IBM(DataDirect OEM)][ODB...

The response from IBM was to reduce the number of objects in the Job Sequence. The only optimization that can be done is for compilation of PX jobs. Whilst I'm sure that Citrixing into the machine is a solution, its not one offered here. I'm closing this as a workaround as there seems to be no magic...

The project has been reindexed and the compilation times have been cut in half.

Still awaiting the report from IBM about other things to do.

The server is offsite so cannot get closer to it

I'll have to find out where the machine is! We dont seem to have any other network issues though.