Search found 15603 matches

by ArndW
Wed Dec 21, 2005 7:45 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Array size
Replies: 3
Views: 1759

The size you choose for your array is dependant upon your allocated DB/2 buffer size. I think that this is 4096 Bytes, the array size should be computed by the following formula INT(<DB Buffer Size>/<record width>) . Choosing your array sizes of 100,000 is guaranteed to provide some incredibly slow ...
by ArndW
Wed Dec 21, 2005 7:41 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Multi Instance Job Triggering
Replies: 5
Views: 1245

My fault in misinterpreting, each job activity stage will wait until it's job has completed before continuing. You would need use a sequencer and have parallel streams within the loop.
by ArndW
Wed Dec 21, 2005 4:43 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Error in populating data from oci to hashfile
Replies: 1
Views: 688

No, this is not a hashed file error. This is an Oracle OCI error due to incorrect configuration or installation. The libraries listed are UNIX type, even though you stated that your installation type is Windows. You do not have the correct PATH setup or have another issue with your installation. It ...
by ArndW
Wed Dec 21, 2005 4:37 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DSD.SEQOpen Failed to open <file> in <dir>
Replies: 12
Views: 9088

Your job is failing because of your DSSendMail routine call; it doesn't seem to have a problem with your sequential source file at all. If you remove your after-job call temporarily (before you fix the parameter errors) the job should run to completion without an error.
by ArndW
Wed Dec 21, 2005 4:33 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DB2 FFTP rows writing
Replies: 3
Views: 921

If you use "." as a column delimiter and have 3 numeric columns, the output will read something like "1.2.3"; if you read that using a column delimiter that is different, let's say a comma ",", then your file has only one column with a value of "1.2.3". Can you split your job into small steps. Get t...
by ArndW
Wed Dec 21, 2005 3:36 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DSD.SEQOpen Failed to open <file> in <dir>
Replies: 12
Views: 9088

Can you open & edit the file in Notepad? If you copy the file in the same directory can you read it through your job if you change it as well? Are you 100% certain that there are no hidden spaces in your filename? You can check this quite easily with the view-data in the designer; even using the...
by ArndW
Wed Dec 21, 2005 2:46 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Packed Decimal Data Type
Replies: 2
Views: 990

reddyds, DataStage supports packed decimal fields. The CFF can do this and a normal sequential stage and use of the sdk functions accomplishes this as well. You will need to explain what your problem is before someone can assist you. Were the packed fields written by PL1 or Cobol? What have you done...
by ArndW
Wed Dec 21, 2005 2:16 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Conditional Lookup
Replies: 6
Views: 1622

Kim - thanks for that! I vaguely recalled that there was something out there, but a couple of searches through the VOC for typical keywords didn't return any results so I though my memory had played a trick on me.
by ArndW
Wed Dec 21, 2005 2:14 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to Select distinct records from Files
Replies: 22
Views: 6264

Kumar,

what is the size limit for a UNIX sort command? I have sorted some large files and it seems to be constrained by the amount of space available for temporary file creation.
by ArndW
Wed Dec 21, 2005 2:11 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: DS Installation
Replies: 4
Views: 1489

At the moment the experts-exchange would be direct competition to this site, as it provides a similar service. The major difference is that this site is not a pay site. I must admit that the experts-exchange has a very good insertion rate into the search engines, many times when I search for keyword...
by ArndW
Tue Dec 20, 2005 5:36 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DB2 FFTP rows writing
Replies: 3
Views: 921

snassimr, this was a bit confusing. If I understand your problem you are writing 8 columns to a FTP stage but only one column is showing up at the receiving end, is this correct? Normally when problems like this occur it is because the column separators are defined differently in the writing and rea...
by ArndW
Tue Dec 20, 2005 5:30 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: (DSOpenJob) Cannot open job ... - not a runnable job
Replies: 5
Views: 6613

Are you calling multi-instance jobs more than once using the same instance name? This is the other common probable cause for this error message.
by ArndW
Tue Dec 20, 2005 5:27 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Split Hash File vs Universe Table approach
Replies: 8
Views: 2913

Dj,

a Universe Table is just another way to look at a UniVerse hashed file.
by ArndW
Tue Dec 20, 2005 5:26 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: RANGE LOOKUP/UV STAGE -READING DATA IN WRONG FORMAT
Replies: 2
Views: 789

My first thought would be that you have specified a precision of 4, which would shift everything over 3 places.
by ArndW
Tue Dec 20, 2005 1:50 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Job Performance for simple server job ?
Replies: 6
Views: 1337

As Ken has already stated, there are too many changes and variables. Find out what your job will do by constraining the output to the DB or changing your target to a flat file. This will give you the max throughput that this job will do. Have you done this, and if so, what is the speed? Now that you...