read delimited: required Column missing - Urgent

vikibemech · Post by **vikibemech** » Sat Mar 27, 2010 6:04 am

My job design is
Seq -> link Partitioner ->Link Collector -> Seq -> Transformer - hash/Seq

I am getting error in 2nd Seq file (CSV file) stage while moving data to transformer. I am getting error message " read delimited: ron 1 "Column name" required column missing.

I have the delimiter in input and output of Seq file stage and both are pipes. This is not the last column.

Below is the error when I changed the attribute to replace.

The Incomplete column value of Column "Column Name"must have values in ascending order.

chulett · Post by **chulett** » Sat Mar 27, 2010 7:28 am

So... you have this error reading the file you just wrote? First triple-check and make sure your file pathnames, options and column metadata match. And undo your 'Incomplete Column' attribute change, it doesn't work that way.

ps. It's not urgent.

vikibemech · Post by **vikibemech** » Sat Mar 27, 2010 7:42 am

Ok chulett. I have checked the path, metadata etc. Everythink looks fine. I have changed the Incomplete column action to Error. Please advice now.

vikibemech · Post by **vikibemech** » Sat Mar 27, 2010 7:45 am

I am able to see the solution provided by you on Tue Nov 17, 2009 10:49 am in other post for the same issue. Since I am not a premium member, I am not able to read completely. if possible could you please update the same in this post also. Thanks in advance!!!

vikibemech · Post by **vikibemech** » Sat Mar 27, 2010 7:51 am

I viewed the data in Datastage and able to see difference in Input (data wrote to file) and output. Input data (data wrote to file) is looking correct and Ouput data seem to improper and junk data.

chulett · Post by **chulett** » Sat Mar 27, 2010 7:53 am

OK. Clarify for us what "improper and junk data" means. Post examples of what you think you are writing out and what "junk" you are finding. Also check from UNIX rather than DataStage, how does it look there?

vikibemech · Post by **vikibemech** » Sat Mar 27, 2010 8:10 am

Improper & junk data - having wrong metadata. Policy number filed is having name of the policy holder.

chulett · Post by **chulett** » Sat Mar 27, 2010 8:22 am

Then you''ve got a delimiter issue. For example, your file is comma delimited, there are commas in your data and you are not quoting strings. That or you're wrong about the metadata.

vikibemech · Post by **vikibemech** » Sat Mar 27, 2010 8:35 am

Chulett, you are correct with first point. Can you explain more on this. I am having delimiter as | (pipe). I have checked both input and Ouput has same demlimiter.

Where to check the quoted string

chulett · Post by **chulett** » Sat Mar 27, 2010 8:49 am

It's set by the Quote Character on the Format tab. Make sure it is set to an acceptable 'quote' character (like ") rather than "000" which means "no quotes". That or strip the pipes from the data if that's allowed / acceptable.

vikibemech · Post by **vikibemech** » Wed Mar 31, 2010 12:56 am

Thanks for the info provided. Issue has been resolved now by filtering policies which has wrong data.

chulett · Post by **chulett** » Wed Mar 31, 2010 7:16 am

Excellent. Please mark the post as Resolved using the button at the top of the page.