Search found 4992 matches

by kcbland
Thu Apr 16, 2009 10:07 am
Forum: General
Topic: How to find the number of users logged into Data Stage?
Replies: 5
Views: 4440

Jobs running should have either an osh process or a phantom process. Do a "ps -ef|grep phantom" to see what's out there running. As for connected Clients, look for dsapi as Craig has shown. You'll have to get the userid's from those processes.
by kcbland
Thu Apr 16, 2009 9:01 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Sequence problem
Replies: 8
Views: 7012

How would the setup impact whether the job was run directly or thru a Sequence? I'm not aware of any difference other than userid and parameters that could be the impact...A trusted connection would be impacted by userid...
by kcbland
Thu Apr 16, 2009 8:37 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Sequence problem
Replies: 8
Views: 7012

Running from a Sequence indicates that a parameter issue may be the case. The Sequence must be running the job with other values. Also, if your Sequence is running as a different userid that would indicate a userid/permissions discrepancy.
by kcbland
Thu Apr 16, 2009 7:26 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Update table directly from a routine
Replies: 3
Views: 3202

We've responded a lot to this question but you are ignoring us. Connecting to a database from a routine requires you to do things not natively supported in the tool. The easiest way to do gain unsupported functionality of communicating with a database is to use the command line gateway ExecSH/ExecDO...
by kcbland
Wed Apr 15, 2009 10:02 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: conncting a database from a routine
Replies: 23
Views: 11953

Return codes from a routine used in a before/after job/stage call aren't easily accessible. If you're calling from a Sequence a DS BASIC Routine you have access to the return code easily and could pass that to the next stage which is a Command stage and uses the command line interface to talk to a d...
by kcbland
Wed Apr 15, 2009 7:56 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Fault type 11 - Abnormal Termination
Replies: 7
Views: 10565

Abnormal Terminations can occur when you do math on NULL values. A reference lookup can introduce NULL values on non-found lookup rows. Look at your expressions in column derivations and constraints and make sure you do NULL handling.
by kcbland
Tue Apr 14, 2009 12:57 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Aggregator performance
Replies: 6
Views: 4471

Look at your cpu utilization - if it's at 100% then it means that you are cpu bound. If it isn't, there's a chance you're paging/swapping memory/disk and could improve performance. Sorting first will help aggregation, but you have to try both ways to see which way the runtime is better.
by kcbland
Tue Apr 14, 2009 11:42 am
Forum: General
Topic: Exporting Project level Director log
Replies: 6
Views: 4231

Every job has its own log file, no matter what a loop thru a list of jobs is required to output all of the logs. Plus, you may only wish a limit on the output from the log. The best idea is to integrate with your job control processes an extraction of interesting log information at runtime followed ...
by kcbland
Tue Apr 14, 2009 10:26 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Routine returning invalid results
Replies: 12
Views: 6969

Take your Routine and paste the logic into a Batch:: style job. Add DSLogInfo() lines to write output to the job log. For developing long routines or functions this can be a useful device. Once you figure out where your logic goes bad, you'll be able to fix your actual Routine.[/u]
by kcbland
Mon Apr 13, 2009 9:20 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Slow Dynamic Hash File
Replies: 10
Views: 6905

Look at the i/o on the filesystem holding the hashed file and make sure it's not running 100%. Install top and watch the cpu utilization (you could look at DS Monitor and look at cpu usage, but that's unreliable. See if the process writing to the hashed file is max'ing out the cpu, if not, look to s...
by kcbland
Mon Apr 13, 2009 5:01 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: dsjob command not working
Replies: 7
Views: 6077

Fully qualify the hostname/userid/password and see if that works...
by kcbland
Wed Apr 08, 2009 12:48 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Delays between Job Sequences / Calling Next Job
Replies: 20
Views: 13004

By saying ALL jobs are slower, we can't easily help pinpoint problems. If we could talk about a specific job then we can focus in on exact issues. For example, if a job runs twice in the same day, processes different volumes, but has longer runtimes for the smaller volumes, maybe we could talk about...
by kcbland
Mon Apr 06, 2009 1:49 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Speed up a hash Lookup (Partial Key) created and build index
Replies: 13
Views: 7754

Your example clearly shows why hashed files should be used in equi-joins as a first and best choice. I suggest you involve a database and use work tables there.
by kcbland
Mon Apr 06, 2009 1:43 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: CHAR to TIMESTAMP
Replies: 1
Views: 2776

Make sure your CHAR data looks like a TIMESTAMP, if not reformat it to look like a TIMESTAMP. If loading into a database make sure generated SQL correctly casts (like with a TO_DATE) else write your own.
by kcbland
Thu Apr 02, 2009 10:09 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: confusion and request for help setup Windows UNIX datastage
Replies: 9
Views: 7587

In case I wasn't absolutely clear - you buy a Windoze license and it only installs on Windoze.