Search found 53125 matches

by ray.wurlod
Fri May 13, 2005 6:35 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Cannot view data from ODBC stage.
Replies: 5
Views: 2057

I think that the problem may, indeed, be a bug. OCONV could be being called by DSD.BCIOpenR (which opens the connection for reading). With the ODBC stage, if a data element is set to Date or Time, then an "implicit" conversion is performed - it's most likely that this uses Oconv (or Iconv, depending...
by ray.wurlod
Fri May 13, 2005 6:31 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DataStage Job 222 Phantom 26691 (Characters not in NLS map)
Replies: 7
Views: 5497

I'd also be asking whether there really are "}" and "|" characters in the data, or whether these are being translated from value marks and field marks.
by ray.wurlod
Fri May 13, 2005 6:27 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: [Help] Fatal error
Replies: 22
Views: 6503

Everyone seems to be bypassing your complaint that you do not have the source, and for a very good reason. This problem cannot be resolved without the source. You will need to reconstruct the job from original specification documents.
by ray.wurlod
Fri May 13, 2005 6:24 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: PX Trainers in India
Replies: 7
Views: 2528

I have been doing some PX training in India recently, but it's for a particular customer and using their own data, so I doubt that they'd be happy about allowing outsiders in. Ascential claim to be able to provide training in India, using a service delivery company in Chennai. For further informatio...
by ray.wurlod
Fri May 13, 2005 6:22 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Can't we use the parallelism feature when we load views
Replies: 5
Views: 1600

I suspect you're right (that you can't select from a view in parallel streams). Check with your DBA or database vendor. Even so, you can still use parallel jobs. Sure you're reading the view as if it were a delimited text file (in a single stream), but that will be partitioned by DataStage into a Da...
by ray.wurlod
Fri May 13, 2005 6:17 pm
Forum: IBM QualityStage
Topic: Split multiple State tokens from single field
Replies: 4
Views: 2910

Pivot stage is a good call. Just make sure that its input link has enough columns to cope with the largest possible list of states. (50 is probably over the top, but is fail safe!)
by ray.wurlod
Thu May 12, 2005 10:24 pm
Forum:
Topic: Reading DataStage Metadata
Replies: 1
Views: 1955

Welcome aboard! :D Yes, you can run queries directly on the Repository, but it would be like putting lipstick on the pig. It adds nothing, and it annoys the pig. Rather than try to get at the actual tables, MetaStage provides two materialized views that vastly simplify the task of getting informatio...
by ray.wurlod
Thu May 12, 2005 9:27 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Urgent: Project has been corrupted
Replies: 19
Views: 4193

SET.INDEX DS_JOBOBJECTS INFORM Record the location of the indexes directory. SET.INDEX DS_JOBOBJECTS TO NULL Now DS_JOBOBJECTS believes it has no indexes. Now effect your repair. fixtool is $DSHOME/bin/fixtool - executed alone or with -H command line option it will generate syntax help. SET.INDEX D...
by ray.wurlod
Thu May 12, 2005 9:09 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to achieve cross multiplication?
Replies: 4
Views: 868

What Kim suggested is exactly what I was going to suggest once I'd clarified that's what you were seeking to do.
by ray.wurlod
Thu May 12, 2005 6:55 pm
Forum: IBM QualityStage
Topic: Split multiple State tokens from single field
Replies: 4
Views: 2910

Can't you just make sure that "-", "," and "." are in your STRIPLIST ?

Not sure why you'd want to force the third character of the first column to be "3", so I'm assuming that the problem there was with the keyboard (or, at least, between the keyboard and the chair). :lol:
by ray.wurlod
Thu May 12, 2005 6:46 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to achieve cross multiplication?
Replies: 4
Views: 868

Please let me clarify. What you're trying to do here is to use metadata (column name) as the basis of a lookup reference key expression. Is this correct?

Any solution to this will be messy, so please verify my understanding before I waste time giving incorrect answers.
by ray.wurlod
Thu May 12, 2005 6:43 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Stage Filter in sequentil file stage command line arguments
Replies: 7
Views: 1733

For starters, the option must be in lower case: sort -u

Secondly, you will almost certainly need to specify at least one sort key (using a -k option).
by ray.wurlod
Thu May 12, 2005 6:41 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: OCI Stage - Data not loaded in a sorted order
Replies: 2
Views: 881

It is a fundamental rule of databases that order of storage is not determinable. The database server makes its own decisions about where to store things. Unless you specify ORDER BY in a query to retrieve those data, you'll get whatever the database server chooses to give you. This is NOT a DataStag...
by ray.wurlod
Thu May 12, 2005 6:37 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Urgent: Project has been corrupted
Replies: 19
Views: 4193

Then again, you might have a corrupted DS_JOBOBJECTS on your backup. Even so you're in a better position to attempt repairs. uvfixfile without the -fix option will tell you whether repairs are necessary. You can then use either uvfixfile or the newer fixtool to effect repairs. However the danger sti...
by ray.wurlod
Thu May 12, 2005 6:30 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: RT_STATUS Error
Replies: 10
Views: 4746

CLEAR.FILE takes a command line argument indicating the table whose data are to be cleared. It was you, in your original post, who claimed to have executed a CLEAR.FILE command. If you didn't specify a table name, then the CLEAR.FILE command would have prompted for one. What name did you provide?!!