Colin
ETL_JOB_HIST,TIME_DIM,ETL_QA_SQL and ETL_ROW_HIST.
Q) How would I go about populating these as Hash Tables ?
You should have the DDL to create these as Oracle tables.
ETL_JOB_HIST should be populated by DSJobReportDb
and so should ETL_ROW_HIST.
You do not need the other 2.
I also noted that when I ran DSJobLinks it converted 260 rows to 87 (a
uniqueness issue).
Q) Is this going to have an impact further down the line ?
This is done on purpose.
ExtractJobStatusSeq and BuildJobStatusHash use a file JobStatusStg.txt.
Q) Where is this generated ?
This is supplied in the zip file in Staging. The Staging folder should be below
the project. It is rebuilt by ExtractJobStatusSeq.
What could I still do if I dont have these tables ?
The whole purpose of these jobs is to load the tables starting with "ETL_".
If you get the email reports running then they are very useful to run
at the end of every sequence. Once you get these working then ETL_QA_SQL
and the QA or Quality Assurance tables can be built and run. This takes
these to another level. This can automatically QA your jobs and email you
the results. Very powerful.
Thanks Kim.
>From: "Larcombe Colin" <
colin.larcombe@fer-ge.ch>
>To: "Kim Duke" <
kim_g_duke@hotmail.com>
>Subject: RE: Your ETLStats generation routine for the Datastage
>Date: Fri, 26 Aug 2005 12:04:04 +0200
>
>Hi Kim,
>
>I have been going through the jobs one by one and converting the Oracle to
>hash files. I started in the following order
> Source Target
>LoadETLJobLink DSJobLinks ETL_LINK (insert)
>LoadETLJob ETL_JOB_HIST ETL_JOB (insert/update)
>LoadETLJob2 ETL_LINK ETL_JOB (insert)
>GetEtlQaSqlStat TIME_DIM ETL_QA_RESULT
>FixedETElapsedTime ETL_JOB_HIST ETL_JOB_HIST (update)
> ETL_ROW_HIST ETL_ROW_HIST (insert/update)
>BuildEtlQaSqlHash ETL_QA_SQL EtlQaSqlHash
>
>I see that there are 4 tables which are not created by any jobs,
>ETL_JOB_HIST,TIME_DIM,ETL_QA_SQL and ETL_ROW_HIST.
>Q) How would I go about populating these as Hash Tables ?
>
>I also noted that when I ran DSJobLinks it converted 260 rows to 87 (a
>uniqueness issue).
>Q) Is this going to have an impact further down the line ?
>
>ExtractJobStatusSeq and BuildJobStatusHash use a file JobStatusStg.txt.
>Q) Where is this generated ?
>
>You mentioned that these tables could be deleted in your previous mail but
>will this not impact how the whole system runs ? What could I still do if I
>dont have these tables ?
>
>I would quite understand if you choose not to continue this conversation as
>it is of little benefit to you and entails, I imagine, a lot of work on your
>behalf. Having said that, I would like to use your system as a database
>agnostic tool.
>
>Thanks
>
>Colin
>