0 votes

I have a shell script which appends a Hive table (overwritten on every run of the workflow) to a Hive archive table. However, on occasion, the first Hive table is empty, which is breaking the workflow because I get an error when the shell script runs:

Failed to access source dataset project_key.dataset_name. Either it is empty or another error occured. Job cannot continue.

Is there a way around this?

by

Please log in or register to answer this question.

1,194 questions
1,226 answers
1,378 comments
11,754 users

┬ęDataiku 2012-2018 - Privacy Policy