Search found 15603 matches

by ArndW
Mon Aug 15, 2005 7:55 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: ODBC Connection to DB/2 - problems with VarChar(1000) column
Replies: 7
Views: 1720

Lance,

tried that as well. I am going to ask the DBAs to give me a view of the table using CHAR to see if that will work, but it will take a day or three to get that done on the mainframe.
by ArndW
Mon Aug 15, 2005 5:52 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Loading DataSets to Oracle Enterprise Stage
Replies: 1
Views: 741

Loading DataSets to Oracle Enterprise Stage

I have a VARCHAR(1020) column which contains normal text plus embedded double-quote characters. When I load this into the Oracle enterprise stage using the bulk loader I get the job aborting because the Oracle data file doesn't expand/change the double quote characters but uses double-quotes as it's...
by ArndW
Sun Aug 14, 2005 9:56 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DELETE.FILE for some files
Replies: 10
Views: 2392

snassimr & Ray, I really had qualms about responding to this question; doing anything directly to the VOC is always dangerous (I've had my bad experiences over the years) and combining a SELECT with a DELETE.FILE was just asking for trouble. I'm glad that snassimr's next post wasn't titled "How ...
by ArndW
Sun Aug 14, 2005 12:30 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DELETE.FILE for some files
Replies: 10
Views: 2392

snassimr, I am not at a UniVerse enabled system now, but I think you can do a 'SELECT VOC WITH @ID LIKE ...TMP... AND WITH F1 LIKE F...' and subsequently do a DELETE.FILE on that select list. You can try the SELECT and do a subsequent 'LIST VOC' to see what you have seelected. The query and delete a...
by ArndW
Sat Aug 13, 2005 10:30 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: job parameters in job control code
Replies: 2
Views: 681

The syntax in the Job Control code should read:

Code: Select all

OPENSEQ input_file TO FILE THEN {...}


You do not need to specify the # hash marks in Job Control code to use parameters, and the quotes you used are also extraneous.

Happy coding...
by ArndW
Sat Aug 13, 2005 7:54 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: concurrent jobs accessing hash lookup for reading &updat
Replies: 2
Views: 869

hiltsmi, the hashed file mechanism can handle many concurrent reads and writes to files, just as any database system does. If one process READs a record that might have been updated by another process (or job), then you need to ensure is that all of your WRITEs and READs are done right off the hashe...
by ArndW
Sat Aug 13, 2005 1:47 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: ODBC Connection to DB/2 - problems with VarChar(1000) column
Replies: 7
Views: 1720

Ray,

just before leaving I found another table which gave the same error - and this one was using DB/2 VarChar(100), once I removed that column the table was readable. I have to find out about VarChar in DB/2 and the wire connect protocol to the host, perhaps I have missed something fundamental.
by ArndW
Sat Aug 13, 2005 1:42 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Kanji Language
Replies: 3
Views: 1225

Vinay, if NLS is telling you that there are unmappable characters then chances are very high you do not have Shift-JIS incoming. Sometimes companies will use Gaiji characters which are not mappable as they are custom characters, but this is rarely used. How certain are you that you Japanese Kanji ar...
by ArndW
Fri Aug 12, 2005 11:36 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: max date of month
Replies: 4
Views: 1277

I don't have access to a test machine right now, but I would take the first day of the next month (using a substring function) and then converting it to Julian and subtracting 1, finally converting it back to timestamp.
by ArndW
Fri Aug 12, 2005 9:10 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to run a job in multiple instances
Replies: 17
Views: 5439

I should have added you can call the job using {YourJobName}.{Instance}, so you just have to modify your scheduler's command line to include a differing instance.
by ArndW
Fri Aug 12, 2005 8:56 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to run a job in multiple instances
Replies: 17
Views: 5439

How about using the system time as the invocation ID, or some other value.
by ArndW
Fri Aug 12, 2005 8:54 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: ODBC problem when migrated from DEV to QA
Replies: 5
Views: 1071

What about your .odbc.ini and project uvodbc.config files?
by ArndW
Fri Aug 12, 2005 8:48 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: ODBC Connection to DB/2 - problems with VarChar(1000) column
Replies: 7
Views: 1720

ODBC Connection to DB/2 - problems with VarChar(1000) column

I am trying to read a table which contains a column declared as VARCHAR(1000) in DB/2. When I remove the column everything works flawlessly. When I try to read the table or even do a view-data on it it aborts with Error executing View Data command: ##E TDOD 000001 14:44:34(000) <main_program> Dataty...
by ArndW
Fri Aug 12, 2005 8:29 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Accessing remote sequential file using local UV Pointer
Replies: 4
Views: 980

In that case your user session and the [background] DataStage job have different access. Try to put a DSLogInfo or DSExecute or other debugging statement in your job where you try to access this drive. I think that is where you are having issues.
by ArndW
Fri Aug 12, 2005 7:50 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: warning due to seq file in parallell EX
Replies: 6
Views: 2152

Is this to a reject link? In that case I don't think you can make the warning disappear; you could put a transform stage between the database stage and your sequential output file, that should work.