Search found 42189 matches

by chulett
Tue Dec 21, 2010 2:23 pm
Forum: General
Topic: Job is not in the right state (compiled and not running)
Replies: 2
Views: 2189

That usually means the job is already running.
by chulett
Tue Dec 21, 2010 10:37 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Upsert in Oracle
Replies: 5
Views: 2608

Technically, you should never "need" hints. As noted, talk to your DBA and have them help you track down the issue. If they don't see anything on their side, you might want to consider involving your official support provider as I seem to recall that it may be an artifact of using those &...
by chulett
Tue Dec 21, 2010 9:00 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Upsert in Oracle
Replies: 5
Views: 2608

First question that comes to mind - is your "ROW_ID" column indexed? Sounds like a full table scan to me, meaning it either doesn't have one or it isn't being used. For the later, involve your DBA and get an explain plan to determine why, typically it turns out to be a 'stats' issue from w...
by chulett
Tue Dec 21, 2010 8:15 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Compile error
Replies: 1
Views: 2013

To me, any kind of apparent "internal limit" error like the one you're seeing from the "storage area manager" and the number of BAPI stages / columns in your job, need to be pursued with your official support provider.
by chulett
Tue Dec 21, 2010 8:10 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: [osh.exe] <defunct> process
Replies: 9
Views: 7653

What operating system? This may help explain things, including why you can't kill them... you can't kill the undead. :wink:
by chulett
Tue Dec 21, 2010 8:06 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Delete data from a table based on four non-key columns
Replies: 4
Views: 2242

Good.

Save the 'custom' SQL for those times when you have no other choice and as a Best Practice let the stages generate the SQL for you whenever possible. Those who come after you will be ever so grateful. :wink:
by chulett
Tue Dec 21, 2010 8:01 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Sparse Lookup
Replies: 14
Views: 9598

You need to help us help you. We have no idea what 'every possible way' you tried and I'm not willing to guess. Did you try the search suggested? What exactly did you try to solve this problem? :? Worst case, we need to know everything you can tell us about your OCI stage. For example - what is your...
by chulett
Tue Dec 21, 2010 1:11 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: how to use DB2 function in stage
Replies: 6
Views: 2582

Sounds more like DDL than DML. :?
by chulett
Tue Dec 21, 2010 12:44 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Delete data from a table based on four non-key columns
Replies: 4
Views: 2242

Why do you feel the need to do anything custom here? Why not leverage the 'Delete existing rows only' update action? That should be all you need.
by chulett
Tue Dec 21, 2010 12:36 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Convert dates to Julian format
Replies: 2
Views: 1346

Rajmohan, I split this question from the completely unrelated post in the General forum that you added it to for some reason. Please take a moment and edit your post to add the relevant details for us - version, O/S, etc. They are only prompted for when you start a new conversation with the Post new...
by chulett
Tue Dec 21, 2010 12:26 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Delete data from a table based on four non-key columns
Replies: 4
Views: 2242

Which DB2 stage? Shouldn't need anything 'custom' for this as there is (typically) a "Delete existing rows only" update action or equivalent available. Don't confuse "key" columns with indexes or primary keys, they are basically just columns to bind into the where clause.
by chulett
Tue Dec 21, 2010 12:18 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Geographical data Type
Replies: 3
Views: 1740

What I've encountered working with the likes of Google and Yahoo is apparently known as "Signed Degrees format" as noted here.
by chulett
Mon Dec 20, 2010 11:41 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Geographical data Type
Replies: 3
Views: 1740

What do you mean here by "geographical datatype"? Something specific to a software package? The geocodes (lat/long) I've dealt with in the past are simple decimal values with a scale of 5 or 6, from what I recall. :?
by chulett
Mon Dec 20, 2010 5:17 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: ODBC error
Replies: 1
Views: 955

First advice for problems that mention the dreaded "CombinedOperatorController" is to add $APT_DISABLE_COMBINATION to your job, set it to True and rerun. The error will (should) still happen but you'll see which stage / operator is actually throwing the error.
by chulett
Mon Dec 20, 2010 5:10 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Sparse Lookup
Replies: 14
Views: 9598

I believe that is only true when there is a key field involved, which (oddly enough) is not required. I recall being able to retrieve a sequence from Oracle by doing a 'select x.nextval from dual' lookup without needing a fake where clause. It would just fire for every row with no input from the sou...