Rangelkup failure message "Updatable lookup requires so

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
Nsg
Premium Member
Premium Member
Posts: 37
Joined: Thu Jan 26, 2006 1:21 pm

Rangelkup failure message "Updatable lookup requires so

Post by Nsg »

Hi all,

I have a sequence where there are 2 jobs -
1) Extract
2) Range lookup and craete output file.

The sequence when triggered, runs both the jobs successfully and creates the file.

However when the sequence is called from other sequence, the Range lookup job fails with the error message of "Updatable lookup requires sorted keys."

The job is doing range lookup on the master link.

This problem seems very strange as the parameters passed as identicle.

Please let me know if you know how to handle this situation.

Thanks,
Nsg
jwiles
Premium Member
Premium Member
Posts: 1274
Joined: Sun Nov 14, 2004 8:50 pm
Contact:

Post by jwiles »

Your job parameters may be identical, but is the environment? Check the values of other environment variables, such as APT_NO_SORT_INSERTION. Maybe they are set by the master sequence.

Regards,
- james wiles


All generalizations are false, including this one - Mark Twain.
Nsg
Premium Member
Premium Member
Posts: 37
Joined: Thu Jan 26, 2006 1:21 pm

Post by Nsg »

Yes, you are correct. I had APT_NO_PART_INSERTION and APT_NO_SORT_INSERTION in master sequencer and setting them to false, makes the job run just fine. Thanks for the help.

The thing I did not know was that these Env_Variables get passed to all the jobs even when not specifically passed.

Is it a good practice to have explicit sorts and partitions in jobs and have APT_NO_PART_INSERTION and APT_NO_SORT_INSERTION set to true in the sequencer.

Thanks,
nsg
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

They're not 'passed' but but they are inherited because they are all running under one environment - the one established by the Sequence. And if your jobs rely on certain values being set, don't assume they are already set in the environment, explicity set them in your jobs. That way a Master Sequence can't come along and override them, all accidental like.
-craig

"You can never have too many knives" -- Logan Nine Fingers
jwiles
Premium Member
Premium Member
Posts: 1274
Joined: Sun Nov 14, 2004 8:50 pm
Contact:

Post by jwiles »

Explicitly defining sorts and partitioning in your job provides you more control over the job's operation and performance, but can be abused if you don't have a good understanding of where and why sorting and partitioning should be done (and NOT done). I prefer it also because it provides a visual indication of where I am performing sorts and partitioning in the jobs I create.

Auto sorts and partitions generally guarantee a correctly-processing job in the great majority of cases, but do not guarantee optimal performance.

Regards,
- james wiles


All generalizations are false, including this one - Mark Twain.
Nsg
Premium Member
Premium Member
Posts: 37
Joined: Thu Jan 26, 2006 1:21 pm

Post by Nsg »

Thanks you all for the explaination and helping resolve the issue.

- nsg
Post Reply