Search found 210 matches

by sbass1
Tue Jul 21, 2009 8:46 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Sourcing environment files from sequence Execute Command stg
Replies: 3
Views: 1954

Thanks Ray. What I'm really looking for is a better understanding of how DS 7.x launches external commands under Unix.

If these details are doc'd somewhere, sorry, please point the direction. I searched on "Execute Command" before posting but got too many false hits.
by sbass1
Tue Jul 21, 2009 8:14 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Sourcing environment files from sequence Execute Command stg
Replies: 3
Views: 1954

Sourcing environment files from sequence Execute Command stg

Hi, I need to execute an external command in a sequence, where that external command needs a number of environment variables set. From what I can deduce, the Execute Command stage executes the external command in the Bourne shell (sh). Since my environment files use the bash syntax, I need to also s...
by sbass1
Tue Jul 21, 2009 7:45 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Merge multiple rows into one based on key values
Replies: 8
Views: 4186

Thanks for the replies. I ran into two issues with this approach: 1) For the sequential files, I wanted to first aggregate the data, as I needed frequency counts per primary keys. If I didn't first aggregate, I couldn't see how to get the frequency counts in the final aggregate stage from multiple i...
by sbass1
Thu Jul 16, 2009 9:38 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Merge multiple rows into one based on key values
Replies: 8
Views: 4186

Merge multiple rows into one based on key values

Hi, I have data from disparate sources (sequential files and database tables) and need to get frequency counts based on the keys. Very simplified example: seq_file --> aggregator --> output_link_1 db_table --> select count(*) from db_table group by key1, key2 --> output_link_2 etc. ... 5 more data s...
by sbass1
Thu Jul 16, 2009 9:23 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Poor performance when updating SQL Server table
Replies: 2
Views: 1176

I've had the DBA create appropriate indexes and the performance is now acceptable.

Thanks for the help...
by sbass1
Tue Jul 14, 2009 11:21 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Poor performance when updating SQL Server table
Replies: 2
Views: 1176

Poor performance when updating SQL Server table

Hi, I'm implementing SCD2 processing against a target SQL Server table. Excerpt from the job: xfm --> Update --> Target Table xfm --> Close --> Target Table xfm --> Insert --> Target Table where Update/Close/Insert links are from the same xfm and to the same Target Table. In other words, one xfm sta...
by sbass1
Fri Jul 10, 2009 3:11 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: New EtlStats soon
Replies: 34
Views: 17611

(I have watched the video and read the install instructions, plus searched on "ETLStats" before posting...) Quick questions about ETLStats: 1) Can I install the jobs into their own project, then report on all the other projects? Or do I have to install the ETLStats jobs in each project on ...
by sbass1
Wed Jul 08, 2009 11:18 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Implementing backdated processing
Replies: 4
Views: 3248

Re: Implementing backdated processing

chulett wrote:Of course you can! Switch the stage to user-defined / custom sql and change the "=" to ">=" just like you've noted and you'll be fine.
Doh! Should have read the help before posting (sorry).

Yep, that fixed it. Thanks for the help!
by sbass1
Wed Jul 08, 2009 11:16 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: SQL Server deadlocks
Replies: 3
Views: 1965

Got no clue how SQL Server does locking, so totally guessing here but... array size? Have you tried setting all links to 1? :? Darn, you're just too fast for me Craig :wink: Yep, I was just about to post that I've solved my own issue. I changed Array Size from 1 to 1000 and Transaction Size from 0 ...
by sbass1
Wed Jul 08, 2009 10:28 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: SQL Server deadlocks
Replies: 3
Views: 1965

SQL Server deadlocks

Hi, This post is related to http://www.dsxchange.com/viewtopic.php?p=332965 So now my job looks like (excerpt): NotBackDated ---------> BackDatedBalance -----> TargetTable (all three links, same DRS stage) BackDatedOpted -------> NotBackDated is "Update or Insert" BackDatedBalance is "...
by sbass1
Wed Jul 08, 2009 1:38 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Implementing backdated processing
Replies: 4
Views: 3248

Use an immediately updated hashed file in the process, to detect whether the record has already been processed (and which will always contain the most recently used value), then Insert or Update into the target. If you prefer, interpose an Aggregator stage to group by key and only deliver the Last ...
by sbass1
Tue Jul 07, 2009 9:38 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Implementing backdated processing
Replies: 4
Views: 3248

Implementing backdated processing

Hi, I have to implement processing where, if a "backdated" record occurs, update a column for all records greater than or equal to the backdate. Say I have this data in my target table: KEY DATE FLAG_VAR 1 20090701 T 1 20090702 T 1 20090703 T 1 20090704 T 2 20090701 F 2 20090702 F 2 200907...
by sbass1
Thu May 21, 2009 11:56 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: row count header to the sequential file
Replies: 17
Views: 6042

I have a source oracle table,transformer and sequentail file. when i run this job i want to see number of records count + all records in sequentail file. row count: 2 column1 column2 1 XXXX 2 XXX Show us 1) exactly what you have (your source), and 2) exactly your desired output (your reformatted so...
by sbass1
Wed May 20, 2009 7:39 pm
Forum: General
Topic: [OT] Help with "vim" on Linux
Replies: 12
Views: 4432

Look for /etc/vimrc, /etc/vim/vimrc, or /usr/local/share/vim/vimrc.

Or do

Code: Select all

find / -name vimrc
, then go get a cup of coffee :wink:

Or create ~/.vimrc, which should override any global settings.

Code: Select all

:scriptnames
should tell you which initialization files are used.

HTH
by sbass1
Wed May 20, 2009 5:43 pm
Forum: General
Topic: [OT] Help with "vim" on Linux
Replies: 12
Views: 4432

Re: [OT] Help with "vim" on Linux

1) It remembers where I was in every dang file from session to session and puts me back there when I vi it again. I would prefer to just start at the top, is there a setting for this? http://durgaprasad.wordpress.com/2007/02/21/remember-file-cursor-position-in-vim-solaris/ http://ubuntuforums.org/s...