Hi all! We’ve identified that a few record updates failed to sync properly around the time our historical data load handed off to our ongoing incremental load. We’ve narrowed down the specific date of the problem, and know that the problem only affects around 0.03% of all records in the table. Unfortunately we’re now a few months down the line from this date, and would ideally like to re-sync (upsert) the records on that date without either: 1. Doing a full refresh 2. Re-syncing all of the updates that have come afterwards (which we might achieve by fiddling with the state) Does anyone know of a way to achieve this? Ultimately we just want to run a one-off “incremental” update with

WHERE updated_at BETWEEN 'START_OF_DATE' AND 'END_OF_DATE'

as an added condition 😄 (PS: We’re loading from PostgreSQL into Snowflake)

You could run a new sync with the starting state that you want that is independent of your

job-id

, such you would simply insert/update the records you're looking for. I think you could even do this on a stream by stream basis with something like:

meltano elt tap-postgres target-snowflake --select my_stream --state '{"my_stream": ...}'

I'm not sure of a way to terminate the tap once it's reached a certain

END_OF_DATE

, though...