An Azure service for ingesting, preparing, and transforming data at scale.
hi there )) thanks for dropping this question on the portal, its super relevant for anyone working with azure tables and adf.
so u got these audit columns in ur azure table right. refresh type and last modified. u want to update them after a full load finishes in adf. U gotta understand how azure table updates work. When u update an entity, u need to include all the properties, not just the ones u wanna change. otherwise, the missing ones will get wiped out. Now about that duplicate columns issue u mentioned. this usually happens when u try to update an entity but u dont include the original values for all properties. adf might treat it like a new set of columns instead of an update.
before u do the update, fetch the current entity from the table. this gives u all the existing values. then, modify only the fields u wanna change (refresh type to incremental, last modified to current timestamp). send the entire entity back with the updates. no missing fields, no duplicates ))
if ur using a web activity or a stored proc in adf, bw sure the payload includes everything. u can also use the azure table storage dataset with upsert, but watch out for those properties. an good example of table operations in adf.
if ur still seeing duplicates, check if u accidentally mapped the same column twice in ur sink dataset. happens to the best of us :)) hope this clears things up.
Best regards,
Alex
and "yes" if you would follow me at Q&A - personaly thx.
P.S. If my answer help to you, please Accept my answer
PPS That is my Answer and not a Comment