DSXchange

Doug, Swap will be used when a page in memory needs to be freed. Normal IBM recommendation for swap is 8gb * number of CPU's. Based on your configuration which I assume is 16 logical nodes (based on your 16 partitions) and at least 3 players per job (possibly more) plus one section leader for each n...

What exactly would you expect to see? If a stream row does not find a match in the reference lookup - what would you reject from the reference lookup? The only thing that can be sent down the reference link is the stream data - that, at least, is known. Flip them and all you have is a bunch of rejec...

Since you indicate that this job has run before and I presume there have been no changes then I would focus my attention on 1) The data has changed 2) The DPF environment has somehow changed. 3) A patch has been installed on the ETL server If #1 and #3 above can be ruled out then I would focus on #2...

Well Kim, I am honored to be recognized as someone that helped and is a friend! We did have some fun at BS and we did learn a bunch! I really do look back on that contract with a bit of fondness - we were all pretty far from home, on a contract that had no chance of success but we still were able to...

Wow... Michael Hester... long time no see.

Every now and again I might be able to answer something :D

Yep - when I first responded it was a parallel thing or so I thought! Then it got moved - dang!

Oh well

The lookup operator will attempt to use memory and when it cannot use memory it will use scratch as defined in your configuration file. If this fills up then it will move to TMPDIR as defined in the environment and if it is not defined then it will use /tmp on Unix and c:\TMP on windoze. If you are ...

If you want to get a sequential count of rows processed regardless of the configuration file used then - @PARTITIONNUM + (@NUMPARTITIONS * (@INROWNUM-1)) will do the trick. If you want the total number of rows sourced then this can work, but an agg with a dummy column will output 1 row. You could so...

>##E TNDB 000000 10:35:38(002) <DB2_UDB_Enterprise_3> Program not bound to database >Attempting to bind to the database /DataStageApp/Ascential/DataStage/PXEngine/bin/db2esql.bnd ....failed Just have your DBA's rebind the package in the DB and you will be good to go. This usually happens after a DB...

Please make sure that in your routine definition that you have added a fully qualified pathname pointing to your .o or .so. If you have then make sure that you have the correct permissions on the library or object. you could put the object/library in the search path by putting it in PXEngine/usr_lib...

Fine - use the filter, but don't do the stage variables.

If the header and 3 detail have the same key you could do a join or merge or even a simple lookup to handle this situation. I would refrain from using stage variables and messing with what node is processing what record etc... not a good way to handle the logic. You could load the headers into a loo...

For future reference, the following message -

APT_CombinedOperatorController

can be further broken down by including the APT_DISABLE_COMBINATION. This will allow the error messages to be more specific. Many of the stages are actually combined operators and this makes diagnosis a bit easier.

In version 8.x there is a new tab of the parallel transformer properties called Surrogate Key. Within this tab you define how you want to manage the surrogate key - either flat file or DB Sequence and then in the transformer derivation of the output column you simply invoke NextSurrogateKey(). If yo...

I think I missed the mark on my previous post. This would work when defined in the schema and read via a stage. Obviously you are past that point and using the modify stage.

Sorry