DSXchange

Don't get into the habit of relying upon the default value of parameters in a parameterset to provide values that can change over time...that will only create the heartache you are now experiencing. The default values are the ones which are only changeable with the recompile of a job. Ray's suggesti...

Good catch! I missed that little typo...might need some new glasses

Regards,

So in the second job (the one which fails, what else does that job do ? Lookups? Aggregations? More joins? More sorts? What is the volume of data (number of bytes) coming into that job (ALL sources)? What does the metadata of your records look like--lots of bounded VarChars with large maximum sizes,...

Two places in the documentation to help jumpstart the thinking process: http://publib.boulder.ibm.com/infocenter/iisinfsv/v8r7/topic/com.ibm.swg.im.iis.ds.parjob.dev.doc/topics/r_deeref_Type_Conversion_Functions.html http://publib.boulder.ibm.com/infocenter/iisinfsv/v8r7/topic/com.ibm.swg.im.iis.ds....

What is the format of the date supplied in the integer?

Regards,

Your class is not defined correctly. More specifically, it does not extend the Stage class, as described in documentation for the Java Pack. Have you read the documentation regarding the Java Pack and Java Client Stages? It describes how to properly create your Java class and programs for use with D...

You asked essentially the same question just last week (except that the quantity was 6 million instead of 3 million) and never answered to the reply given. Here is the post: http://dsxchange.com/viewtopic.php?t=148486 In addition to the questions asked of you in that post: - Is /InformationServer/Sc...

To answer your question: Sometimes that might be the best way. Before modifying the job(s), review the following items: 1) How much space does the data consume? # records * record length = # bytes. Calculate for each sort! 2) Are the columns primarily bounded varchar--that is, varchar with a defined...

If the data you wish to use in the second job can be created purely within the source database through joins/etc (i.e. no DataStage processing required), you or a DBA could possibly create a View which would return those columns and data. If not, such as when some of the columns are created or manip...

Oracle NUMBER fields (with no precision/scale defined) are imported into DataStage as Decimal(38,10) by default. The import process has no visibility to the actual data values present. You are certainly free to modify the scale value in the Table Definition you imported if you desire.

Regards,

Only if you need to save the decimal portion of the value.

Regards,

Yes, thank you indeed!

Regards,

Load Leveler documentation is available in the online IBM Cluster Information Center (a quick Google search will get you there). Other than the AIX-specific items (if any) for installing Load Leveler, there will be very little difference in LL<->IS configuration from what is listed in the Redbook yo...

My mistake...typing it in from memory :( Index('|ARC|PPD|CCD|CTX|COR|DNE|','|':lnk_xfmr_target.CLASS:'|' ,1 ) The ,1 is the third argument (missing from my original). You can also just click the [...] button in the derivation editor and search for Index in the String functions. Selecting it will pro...

As Craig states, #2 is incorrect. You are guaranteed that at least one of your conditions within the parenthesis will always be true. You might as well have "AND 1=1" instead! The equivalent Index function would be AND Index('|ARC|PPD|CCD|CTX|COR|DNE|','|':lnk_xfmr_target.CLASS:'|') >= 0 w...