05-31-2017 08:32 PM
What's the best practice to trigger Oozie workflow based on availability of data in external system (e.g. RDBMS)? The requirement being that the workflow should start as soon as (or at least very soon after) the data in the external system is available.
One approach is to schedule the job with high frequency, check for the data and exit if it is not present. But this doesn't seem particularly elegant. Is there any other way?