DSXchange

Hi, It seems that in "Upsert", DataStage does update first and if not found then insert but I also faced the similar problem. In fact Datastage does an "Insert" first. So duplicate records may exists in this case. So to avoid this situation you have to handle it by Database uniqu...

Hi,

Please give the data that you expect. I want to compare it.

Regards
Amitava

Hi,

Can you give some example of the output data ?? Also check the DataStage parameter APT_STRING_PADCHAR.

Regards
~Amitava

Hi,

DataStage takes care about Null automatically in Sort stage.

Regards
Amitava

Hi,

There should not be any problem in this design. Check 2 things. If Z is a nullable column then handle null before join. Other is check the partitioning. Better if you use hash partition on the joining keys.

Regards
Amitava

Hi,

Check the Osh definition of your job. You can find it by Job Properties -> Generated Osh. This definition is not in sync with your job design.

Regards
Amitava

Hi,

You have to construct the file name with path in File stage. Check the online manual.

Regards
Amitava

Hi,

Set the "Nullability" of the field to "Yes"

Regards
Amitava

Hi,

If your primary key of the table is a running sequence then you have to modify your insert SQl that you are using in datastage (Check the datastage manual for Oracle insert). Otherwise you can use a "Upsert" option for Update/Insert operation.

Regards
Amitava

Hi,

Give some more information like your job design, partition configuration, APT_Config file etc.

Regards
Amitava

Hi,

You can't compile a shared container. You can only save it.

You just try to attach the shared container in your main job and define the input and output links of the container along with table definitions very carefully. Then compile the job. I think it will work.

Regards
Amitava

Hi,

You can use this

If IsNull(TMMBR_ADJ_A) = 1 Then <Some default value or Space(<Field Length>) Else If TMMBR_ADJ_A > 0 then 'Y' else 'N'.

Regards
~Amitava

Hi,

DataStage has incorporated a very improved and flexible stages to handle job sequencing activities in 7.5 including Restart logic. You should use this. After all, Basic routines are not parallel.

Regards
Amitava

Hi, If you use a user defined query in DB2 enterprise stage, Datastage works in sequential mode. Instead of using user defined query use the "Read Method = Table" and use the "Table", "Select" and "Where" clauses under this. Then it will work in parallel. In l...

Hi,

Datastage left outer and right outer join works like database. If you change the link ordering, it will work differently.

Regards
Amitava