Search found 22 matches

by christy
Fri Feb 08, 2008 4:02 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Add Multivalue Counter into the output
Replies: 1
Views: 1053

Add Multivalue Counter into the output

I have a UV (UniVerse) Stage as input. The input data is normalized on the association "detail". The input data looks like: ID: 100 Status: E@VMD@VME@VMR... Code: 12@VM34@VM12@VM12... ID: 200 Status: E@VMF Code: 74@VM12 the output looks like: ID: 100 Status: E Code: 12 ID: 100 Status: D Co...
by christy
Thu Jan 18, 2007 12:34 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: The connection is broken (81002)
Replies: 14
Views: 8041

I am having the same problem again today. I edited a job and saved it (I was able to open DS designer). When I went to compile, I got an error as if there was a syntax error compiling. So I changed it back to where it was before and tried to compile. Same error. I closed DS designer. I then would NO...
by christy
Wed Jan 17, 2007 5:48 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: The connection is broken (81002)
Replies: 14
Views: 8041

After another reboot (the 5th try), I finally got it to work. I recompiled all the jobs and things seem to be working ok now.
by christy
Wed Jan 17, 2007 3:24 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: The connection is broken (81002)
Replies: 14
Views: 8041

Thanks. I had been searching the forums for 81002 for quite sometime. The reboot and REINDEX worked before. I tried COUNT DS_JOBS and DS_JOBOBJECTS and they both returned correct results. It eorrs when I try to open everthing DS related: Administrator, Manager, Designer or Director. They are all ins...
by christy
Wed Jan 17, 2007 3:03 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: The connection is broken (81002)
Replies: 14
Views: 8041

Thanks for the reply.... I don't think it is a network issue because I am working from the server and the SQL database is on the same server as DS jobs. I dont see the need of doing a REINDEX of your repository just because you are having network issues. Search for keywords "the connection is b...
by christy
Wed Jan 17, 2007 2:54 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: The connection is broken (81002)
Replies: 14
Views: 8041

The connection is broken (81002)

In the past when I got "the connection is broken (81002)" error, I would re-start the machine; then ran "DS.REINDEX ALL" in Administrator; then open DS designer; everything would be fine. However, today when I got the error, I re-started the machine and ran "DS.REINDEX ALL&q...
by christy
Mon May 22, 2006 5:02 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: syntax for key matching
Replies: 6
Views: 1979

Code: Select all

Field(SO_ID,"*",1)
I am afraid that won't do it either. That only work if the detail is the primary input. In my case, the header is the primary input. SO_ID is 12345. the detail key SODET_ID is 12345*001. The Key express needs to be for the detail.
by christy
Mon May 22, 2006 4:56 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: syntax for key matching
Replies: 6
Views: 1979

aditya wrote: DSLinkXX.SO_ID:"*":DSLinkXX.LINE
I won't be able to use that because both links are input links. I would have to say something like that in the Key Expression:

Link1.SO_ID:"*":Link2.Line

It didn't seem to like it. I got no return.
by christy
Mon May 22, 2006 4:16 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: syntax for key matching
Replies: 6
Views: 1979

syntax for key matching

The Primary input is sales order, key is SO_ID (12345) 2ndary input is sales detail, key is SODET_ID. The SODET_ID field is like "SO_ID*Line" (12345*001) How do I specify the Key Expression for SODET_ID of the 2ndary input in the transformer? I only know how to do it if the sales detail is...
by christy
Tue May 16, 2006 2:36 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: table join and normalization
Replies: 5
Views: 1851

Thats true. But it depends upon how much data are we looking at. I, personally have not done inner joins using UNEST and IN, i do it the old school way, by writing user defined queries. Regards, The subquery is a user-defined query. The subquery part (the inner select) should eliminate a lot of rec...
by christy
Tue May 16, 2006 1:48 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: table join and normalization
Replies: 5
Views: 1851

Try doing the same thing with User Define Sql I copy and pasted with User Defined SQL option. It didn't work. However, I rewrite the query to be a subquery: SELECT T1... FROM UNNEST T1 ON T1_DETAIL WHERE T1.key IN (SELECT T2.key FROM T2 WHERE T2.F1 = "A") That seems to be working ok. I am...
by christy
Tue May 16, 2006 12:14 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: table join and normalization
Replies: 5
Views: 1851

table join and normalization

This is a UV stage. I have two files/tables T1 and T2. T1 is normalized on T1_Detail. I am using "Generated Query" option. In the table names box, I entered: T1,T2. In Columns tab, all columns are from T1. In Selection tab, I entered: T1.key = T2.Key AND T2.F1 = 10 When I view the query, t...
by christy
Tue May 16, 2006 9:22 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Len Function
Replies: 2
Views: 1398

DSguru2B wrote:If you are using it in the where clause then use the following

Code: Select all

CHAR_LENGTH(table1.Field1) = 10
That should do the trick.
Cheers.
Thank you so very much. Exactly what I needed...
by christy
Tue May 16, 2006 9:10 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Len Function
Replies: 2
Views: 1398

Len Function

I am trying to use the len function in the where clause under "selection": LEN(table1.Field1) = 10 to limit the number of records returned from a universe database. I am getting an error: UniVerse/SQL: syntax error. Unexpected symbol. Token was "LEN". It didn't seem to like the L...
by christy
Fri Mar 03, 2006 4:52 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Hashed file performance
Replies: 1
Views: 1302

Hashed file performance

Here is the scenario: all jobs run once every night. job1 creates a hashed file: h1 job2 makes updates to the same hashed file: h1 job3, job4 and job5 need to read from the hashed file: h1. The hash file h1 has about 15,000 records. Q1: I took the defult settings for creating h1 in job1. Is it the m...