Search found 42189 matches

by chulett
Mon Feb 22, 2010 12:18 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Exit status for orchadmin delete command
Replies: 5
Views: 1555

From what has been posted here, that's just The Way It Works. Perhaps you can parse the Output variable and look for failure keywords?
by chulett
Mon Feb 22, 2010 12:16 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: odbc stage data loading slow
Replies: 3
Views: 1597

What kind of 'load' are you doing? Perhaps what was inserts before are now all updates.
by chulett
Sun Feb 21, 2010 6:04 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Accesing Variables in Sequences
Replies: 2
Views: 1127

Use a Server job and put the id into the job's USERSTATUS area for easy access. Search here for DSSetUserStatus for examples of the wrapper routine you'll need for that.
by chulett
Sun Feb 21, 2010 6:02 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Aggregation - help with logic
Replies: 15
Views: 3863

Or that. :wink: :lol: Forgot I've done 'conditional counts' the same way - add a new field and conditionally set it to 1 or 0 (depending on if it should be 'counted' or not) and then sum the field. Typically this happens when a value could be conditionally counted in several groups and the counts be...
by chulett
Sun Feb 21, 2010 6:01 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Any limitation on size when using Iconv and Oconv with "
Replies: 29
Views: 7641

I had problems with MOD() on numbers over a certain size and I didn't see a need to bump up the EXACTNUMERIC value so I wrote an "SMOD" routine that used the string math functions I mentioned earlier to do a proper mod on any big ass number. Worked fine for me. Would you describe these, t...
by chulett
Sun Feb 21, 2010 12:00 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Aggregation - help with logic
Replies: 15
Views: 3863

Something is popping up in the back of my mind re: your SumIf situation. From what I recall, you can set up the Aggregator to do sums and then one of the 'sum' options is 'Count Non-Null occurances' in a certain field. I'm paraphrasing but that should be close. It can get a little tricky as there ar...
by chulett
Sun Feb 21, 2010 11:31 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Any limitation on size when using Iconv and Oconv with "
Replies: 29
Views: 7641

BASIC and all documented in the BASIC manual. So, rather than "div" to divide you would use "sdiv". Same for all of the other operations: sadd, ssub, etc. Build your own mod function that just uses string math.
by chulett
Sun Feb 21, 2010 10:52 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Aggregation - help with logic
Replies: 15
Views: 3863

I don't have access to any kind of DataStage server at the moment, so can't help with the sum/count issue but I remember thinking I had to do them separately as well but then could swear I managed to do both at once. Others will have to help with that. For the pivot, the aggregator output will be mu...
by chulett
Sun Feb 21, 2010 10:08 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Aggregation - help with logic
Replies: 15
Views: 3863

A little lost but the statement that the aggregator "cannot handle sum and count at once" is not true from what I recall. The types make it seem that way but I could swear I was able to combine the two when I needed to. :? That and your "SumIf/CountIf" functions are simple sums a...
by chulett
Sun Feb 21, 2010 9:46 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: new log file needed for each job/sequence execution
Replies: 4
Views: 1165

Your script should just ask the job directly using dsjob -jobinfo to get the status of the most recent run, unless you need to look back further than that? :?
by chulett
Sun Feb 21, 2010 9:44 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Datastage log through Unix
Replies: 3
Views: 1189

Not all by itself as it works on one specific job at a time. People will typically use another option to get a list of all job names in a project and then loop through that list. There is also an option to get all projects.
by chulett
Sun Feb 21, 2010 9:17 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Datastage log through Unix
Replies: 3
Views: 1189

Sort of, the dsjob command line function has options to pull out log information: -logsum, -logdetail, -lognewest specifically. These are documented in the Command Line Interface chapter of the Server Job Developer's Guide pdf, amongst other places.
by chulett
Sun Feb 21, 2010 9:12 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Any limitation on size when using Iconv and Oconv with "
Replies: 29
Views: 7641

I had problems with MOD() on numbers over a certain size and I didn't see a need to bump up the EXACTNUMERIC value so I wrote an "SMOD" routine that used the string math functions I mentioned earlier to do a proper mod on any big ass number. Worked fine for me.
by chulett
Sat Feb 20, 2010 10:48 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Encrypt DB Password using Oracle Wallet
Replies: 5
Views: 2354

Sounds like a Known Workaround to me. :wink:

After you create the encrypted passwords in the Administrator, be sure to change their default value to $PROJDEF after you add them to any job.