Search found 53125 matches

by ray.wurlod
Mon Dec 15, 2008 9:50 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: how to copy hash file from one folder to another folder
Replies: 1
Views: 785

You have marked your job type as Parallel. I am assuming that that is incorrect. It depends (on whether the hashed file is dynamic or static) but, since it's probably dynamic, then you need a different process. The "hashed file" - and please note the terminology - it's not "hash"...
by ray.wurlod
Mon Dec 15, 2008 9:43 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: generating schema file
Replies: 10
Views: 3714

Nothing shown in this thread is a DataStage schema file.

You can save schema files directly from the Table Definition record (in Manager or Designer in versions earlier than 8, in Designer in version 8 or later).
by ray.wurlod
Mon Dec 15, 2008 9:40 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: APT_BadAlloc: Heap allocation failed
Replies: 9
Views: 9714

HOW big is the row (in bytes, not in columns)? You will find the size of each data type documented in the manual.

Once you know the row size, THEN you can calculate the optimal array size.
by ray.wurlod
Mon Dec 15, 2008 9:37 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: best practice to handle billion records
Replies: 3
Views: 1730

1. Search DSXchange for ideas about throughput performance. 2. Create a job that identifies which rows are to be inserted and which have to be updated/deleted. Bulk load the inserts. 3. Create multiple instances of your job each of which processes a proportion of the records. (Be very careful with t...
by ray.wurlod
Mon Dec 15, 2008 9:34 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Input column does not promote to output column
Replies: 5
Views: 4416

Either way, post the data types on the input link and the data types on the output link.
by ray.wurlod
Mon Dec 15, 2008 9:32 pm
Forum: General
Topic: Converting Date using regular transformer
Replies: 3
Views: 1506

Welcome aboard. You won't get a timestamp without specifying a time part. You won't get proper handling of null unless you handle the null. For example If IsNull(InLink.TheString) Then SetNull() Else ... Given that the date is a string use string operators (substring and concatenation) to construct ...
by ray.wurlod
Mon Dec 15, 2008 9:29 pm
Forum: General
Topic: Unable to delete datasets using orchadmin
Replies: 2
Views: 5048

This is basic, fundamental, operating system 101 stuff. It is totally unrelated to DataStage or Information Server. "Command not found" means that the operating system can not find the command, which means that the command's parent directory is not included in the command search list - tha...
by ray.wurlod
Mon Dec 15, 2008 9:26 pm
Forum: General
Topic: UniVerse RetrieVe syntax for accessing field number
Replies: 2
Views: 1454

Why do you believe the long description is in field #7? I suspect (without checking) that the long description is usually in field #5 (except for DS_JOBS). Anyway, to answer the question, you clearly have not been looking in the right places. Any particular field number can be extracted using the Fi...
by ray.wurlod
Mon Dec 15, 2008 9:21 pm
Forum: General
Topic: Unable to write to datasets
Replies: 6
Views: 1512

Are you sure? The error message refers to hostname z2dquat1.
by ray.wurlod
Mon Dec 15, 2008 3:40 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: OCI Stage - Issue with reading decimal values
Replies: 11
Views: 3458

What is different between production and development? ("Nothing" is not the correct answer.)
by ray.wurlod
Mon Dec 15, 2008 3:39 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Row Generator & Year Date
Replies: 8
Views: 2882

Perhaps you can now mark this thread as Resolved using the green button at the top?
by ray.wurlod
Mon Dec 15, 2008 3:37 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Complex file reading
Replies: 8
Views: 2350

Add a reject output link from the Sequential File stage; this will capture any row that does not match the record schema for a detail record, and downstream stages on this link can re-parse for example to retrieve the dates and totals. record schema { policy_no string[8]; filler1 string[1]; table_no...
by ray.wurlod
Mon Dec 15, 2008 3:31 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Error at login - Failed to create the ASB COM wrapper object
Replies: 7
Views: 4489

The installation is probably OK. ASB in this case is "Application Server Backbone", the layer provided by the WebSphere Application Server to deliver the services used by products such as DataStage to communicate with Information Server. I'd begin by verifying that the WebSphere Applicatio...
by ray.wurlod
Mon Dec 15, 2008 3:28 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: record dropping
Replies: 4
Views: 1490

One hour left for what? DSXchange is an all-volunteer site whose members post as and when they can. You posted at 2am my time, there's no way I'm awake at that hour. If you need urgent help, sign up with your official support provider for premium service, and learn just how much "urgent" c...
by ray.wurlod
Mon Dec 15, 2008 3:25 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Splitting single column values(length not fixed)
Replies: 5
Views: 1571

Use a Column Import stage to do the parsing for you. Upstream the "first line is column names" (so ignore the first line) is asserted. Or trap that on the reject link of the Column Import stage.