Avoid Warnings

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
Django
Premium Member
Premium Member
Posts: 20
Joined: Fri Apr 18, 2008 12:16 am

Avoid Warnings

Post by Django »

I used to work with Server Edition. One of my unit test tick off was to not have any warnings at all. Now I'm working in 8.0 Information Server using parallel Jobs. Cant have a job without warnings. Is it possible to do something about this ?

So many types of warnings

Eg.
SA_STORE: Column NET_RETAIL floating point decimal is not fully supported; adjusting the scale.

SA_STORE: When checking operator: When binding output interface field "Store" to field "Store": Implicit conversion from source type "int64" to result type "int16": Possible range limitation.

Sequential_File_3: When checking operator: When validating export schema: At field "Net_Retail": "null_field" length (1) must match field's fixed width (14)

Lookup_39,0: When binding input interface field "Dept" to field "Dept": Converting a nullable source to a non-nullable result;
a fatal runtime error could occur;
use a modify operator to specify the value to which the null should be converted.

This story is not just one Job in most or all of my jobs. I deasire not to have any warnings .

Any idea ?
Django
If a simple dsx project requires such intelligence how much more is required to create this Cosmic System. Who dares to say there is no intelligence behind this creation i.e. no GOD .....
telenet_bi
Premium Member
Premium Member
Posts: 33
Joined: Wed Jul 23, 2008 7:33 am
Location: Mechelen, Belgium
Contact:

Post by telenet_bi »

there's 2 ways of solving this:
- change the design to do what the warnings say (example:use a modify operator to specify )
-suppress the warnings from the log with the message handler.(check the manuals, I'm not sure where it is.) you can here change the type (warning becomes info) or you can suppress it completely from the job-log.

we for instance have suppressed the first warning you give: typically if you do a calculation in a userdefined Oracle query you get these floating point decimal errors.
converting non-nullable to nullable is something where I feel it's better to do this explicitly in stead of suppressing this warning.
Post Reply