DSXchange

The thing is I need to continue with the value from last job.

How can I implement it, very confused.

Somebody can paste a link for me ?

I know there is a SDK API routine KeyMgtGetNextValue.

But it looks like illegal for parallel job, anybody can help ..

Thanks !

And I will try it

I need to add one row head and one row tail to my output data file. And I use funnel to combine 3 different inputs. Input1 is the one row header, Input2 is the body of data file Input3 is the tail of the file. I use sequence mode for funnel. partition use default. I guess it is a partition problem, ...

I think I sort the key columns for aggegator, and still get the warning..

I got a new warning in my job log. It is

"Hash table has grown to 16384 entries."
"Hash table has grown to 32768 entries."

Anybody knows how to get rid of them ?

Join_AXIS_ADDR.CIDNAXISSRVC_Sort: When checking operator: Operator of type "APT_TSortOperator": will partition despite the preserve-partitioning flag on the data set on input port 0. Anyway can help, what option I need to choose to get rid of the warning, and guarantee the join result is s...

My case is like id1, id2 1, a 1, a 1, b 1, b 1, c 2, a 3, a 4, c I need to remove any id1 who has more than 2 different id2. It is a little bit tricky. What I did is remove duplicate based on id1, id2, hope to get id1,id2 1,a 1,b 1,c 2,a 3,a 4,c Then I use aggregator to count rows based on id1, hope...

Update about the problem, I am still working on it and finding some hints If i put a remove dupliate stage in front of aggregator, the output result is what i want ? And if I use aggregator stage inside unique option for the input stream, and the output of aggregator is not right, there is one row o...

I spend around 3 hours and use change capture stage and find the problem. I have some duplicate keys (actually only 1 row) in my right input , which is supposed to be unique. Two strange things. 1) I used remove duplicated stage on the right input, and it didn't remove out the duplicated one, but it...

Yes, I use hash partition and sort by the join key column

The very strange thing is that if I choose unique for the right link in sort stage, it is right.

but the right input link is already removed duplicated by remove_duplicate stage.

Yes, I use hash partition and sort by the join key column

The very strange thing is that if I choose unique for the right link in sort stage, it is right.

but the right input link is already removed duplicated by remove_duplicate stage.

I got a very strange problem today from join stage. Use inner join left input, Dataset A has 5M rows, copy dataset A to aggregator, group by keya,( count rows, and I need to filter out some of them) Output to right dataset B, which has 830 rows, and I didn remove duplicate, still 830 rows. Inner joi...

I use aggregator to count rows, and count rows value is stored in a field callded countOfRows I assiged the datatype of countOfRows as integer, decimal, numeric, double, float, Datastage always pops up warning like "When binding output interface field "RecNO" to field "RecNO"...

I want to use modify stage to transfer columna to columnaa by remove the last bit of columna, and convert to number, basically Columna like 6572888. varchar2 Columnaa should be 6572888 number. And i use Columnaa:decimal[10] = decimal_from_string(substring[0,10](Columna)) I guess my questions is whet...