Search found 15603 matches

by ArndW
Fri Feb 02, 2007 3:34 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to take the splitted records as seperate records
Replies: 4
Views: 1277

Write it to a sequential file output and then read it with the new format, but declare this file to be a named pipe so no I/O is done.
by ArndW
Thu Feb 01, 2007 10:49 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Reading file with variable record length
Replies: 13
Views: 4561

Derek - Use a REDEFINES in your COBOL FD, then force string truncation in your job by using the substring of the first n-bytes from your string length field.
by ArndW
Thu Feb 01, 2007 10:47 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Data Set Management is very slow!
Replies: 20
Views: 4288

How long does it take for you to dump the dataset into a flat file from the command line? (orchadmin dump)?
by ArndW
Thu Feb 01, 2007 10:39 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Version Control: How do you do at your place
Replies: 18
Views: 4068

Ernie - I've never liked V8 too much, prefer something higher octane such as rum :lol:
by ArndW
Thu Feb 01, 2007 10:37 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Viewing records in error, from log
Replies: 6
Views: 1665

If you look up the topic of capturing rejected rows on write you will see that you can store the error messages with the rows that caused them. The log files (as you've already noticed) don't always show that.
by ArndW
Thu Feb 01, 2007 10:35 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Data Set Management is very slow!
Replies: 20
Views: 4288

splayer - <sigh> So now your slow dataset is gone; kind of hard to analyze what could have caused performance issues now that it is gone. Did you check the orchadmin command line "ll" prior to deleting it to see if the delay was there or in DS or if any error messages were produced in the command li...
by ArndW
Thu Feb 01, 2007 4:12 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Lookup failure
Replies: 3
Views: 1031

Vinod - ok, the lookup fails and doesn't find a value. What is your question?
by ArndW
Thu Feb 01, 2007 2:57 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Data Set Management is very slow!
Replies: 20
Views: 4288

splayer wrote:I executed the dsenv from my /home/myid directory by doing:
../dsenv


Note you need to execute ".[space]./dsenv"
by ArndW
Thu Feb 01, 2007 2:42 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Error in assignment. SQLSTATE=22005
Replies: 4
Views: 15058

What are you datatypes for key1 and key2 and what is the datatype in your DS job with which you are filling them? For those 2 values of key1 and key2 what happens when you try to manually type in the SELECT clause from the command line and the DB2 command?
by ArndW
Thu Feb 01, 2007 2:34 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: '##' Parameter
Replies: 6
Views: 1703

suresh - if we take a step back, you are saying that when you use text values for a job parameter it works, but if you use two hash signs it does not work? Since the hash mark, '#', gets interpreted by a UNIX shell to have a wildcard meaning and is also used in DataStage in some places to mark a par...
by ArndW
Thu Feb 01, 2007 2:21 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: send a file with $ 0.00 amount to bank
Replies: 2
Views: 893

Welcome to DSXchange, preddy! I understand the business logic from your description in the post, but the approach you are going to use to send data for new parties is still vague. Can't you select all members from the party table where there are no corresponding payments in pmt_lin to get those valu...
by ArndW
Thu Feb 01, 2007 2:15 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Reading file with variable record length
Replies: 13
Views: 4561

Derek - are the first 2 bytes of the record that denote the length actually binary or display (as shown in your original post example data)? I used DISPLAY and was having issues, but think that S9(4) binary would work.
by ArndW
Wed Jan 31, 2007 3:47 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: How to Convert ASCII CHAR into ASCII BINARY format
Replies: 13
Views: 4130

Correct, SEQ() only returns one number. You will have to write a loop around that function.
by ArndW
Wed Jan 31, 2007 12:30 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Data Set Management is very slow!
Replies: 20
Views: 4288

the dsenv is just a script, copy it to your home directory and ". ./dsenv" execute it there.
by ArndW
Wed Jan 31, 2007 12:27 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Lock and unlock jobs
Replies: 20
Views: 5670

Job status cannot be set per user. Changing OS-level permissions on some of the part files won't quite do it either, in addition they will generate errors and warnings for users just browsing the repository.