DSXchange

Download EtlStats from my tips page.

This maybe be true. Vincent made a change to EtlStats to aggregate the nodes into one number to get the true throughput.

Ken

Edit your post and change @FM to @VM.

Awesome blog.

Several us started with the beta version 1.0. Some before that. That was back in 1997. VMark was the company that originally wrote DataStage. VMark made their money off a database called uniVerse. Many of us were employees of VMark. I have been using uniVerse since 1986. There are several good posts...

I have not tried this but it should give you the field number: SELECT DS_JOBS.NAME AS JOB_NAME, DS_JOBS.CATEGORY, DS_JOBOBJECTS.NAME AS OBJECT_NAME, DS_JOBOBJECTS.OLETYPE, EVAL DS_JOBOBJECTS."if index(upcase(@RECORD),':1',1) = 0 then 0 else dcount(@RECORD[1, index(upcase...

By the way DwNav only brings in sequences to get the relationship between sequences and the jobs they run. It always annoys me to try to figure out what sequence controls the job I need to change. Usually you have to do a search like the above which I have a cool routine JR Hines wrote or I go to pr...

It is easier to do this at TCL. You need to search for TCL commands. TCL is the way to talk to Universe directly or what is now called the DataStage Engine. You get a TCL prompt ">" when you telnet into a DataStage server. This is equivalent to sqlplus for Oracle. The quickest way to find these reco...

You are correct. This is probably not available without hacking DS_JOBOBJECTS. Not a recommended practice for a novice. I am sure one of the top posters could do it in a few minutes if you could convice them it was worth their effort. Not sure how you can do that.

Number 1 would be nice.

The status of DSJ.ME should always be "Running". If you look at job control generated for sequences. Then you can see how to get job status. You may need to leave off the job run and the wait for job routine calls but you need to attach to a job first. There is lots of code posted by me and others w...

How are you creating partitions? Distributed hashed files is very high level Universe concept.

I assumed you used a UV stage. If so then post the SQL. If no then what hashed file are you reading from?

StageName is really StageType or LinkType. I doubt if things printed like you think it did. The problem is it printed down the page because the columns are too wide for a page. Eliminate some columns.

EtlStats has a script named DsJobLinksAll.sql. Here is the SQL: SELECT DS_JOBOBJECTS.OLETYPE as LinkType, DS_JOBS.NAME as JobName, DS_JOBOBJECTS.NAME as LinkName, DS_JOBS.CATEGORY as Category FROM DS_JOBOBJECTS, DS_JOBS WHERE OBJIDNO = JOBNO and DS_JOBOBJECTS.OLETYPE like '%put' Group by LinkType, J...