DSXchange

Hi

I am joining two sources(a file and a table) with one key and if there is no match, then i am joining the non-matched records with some other field of the table.

I am just using Auto-partition throughout the job.

The first join is working properly..but the second join is not..If it is a patrition problem it should affect both isnt it..

Hi,
How are collecting non matching records out of a join? Is there any provision in the Join stage to collect them? Please be more clear.

Thanks.

Try running your job with a 1-node configuration and then with a 2-node configuration. If the results are different, then you have a partitioning problem that you need to correct.

have you checked the key checkbox?
can you use hash partition?

The keys you partition on in the first join will be different from the keys you partition on in the second join.

What are the various stages in your job?

I made the second join to run in sequential, but still some of the records are not joined properly.

Did you check the datatype, length etc of the joining columns ?

Is it a case sensitive match ?

What do the records which do not join have in common? Try to locate the keys of missing records in the counterpart source manually.

It is a varchar field of length 13 on both the links. When i query it from the SQL i am getting an output, but when the join happens it doesnt match..It is a case insensitive match only.

Some control characters might be appended to data in the database.
Use rtrim , ltrim on the data present in the database.

Then run the job.

Regards
Sreeni