Record count mismatch in DS and UNIX

nynali · Post by **nynali** » Mon May 21, 2007 9:21 am

Hi,
I have DS job which does a LEFT OUTER JOIN b/w 2 sequential files.
This job reads the files NDMed to UNIX in Binary mode and does a sort of both files on KEYS and does a join.The job went fine and it showed the output file(SEQUENTIAL STAGE) count as same as the left input file which should happen in our case.
The output file RECORD options were given as
FINAL DELIMITER=UNIX new line
RECORD LENGTH = FIXED.
DELIMITER = NONE.
But when I do a wc -l in UNIX on the same output file the record count is more(not double).I have NDMed the same file onto Mainframes and came to to know that some of the records in between are written in two lines instead of one.
Please suggest on this.

ray.wurlod · Post by **ray.wurlod** » Mon May 21, 2007 2:51 pm

The obvious thing to check for is newline characters in your data.