In our development environment, I see that looker creates multiple versions of same PDT with different hashkeys (Maximum up to 6).
I have many PDTs that that are triggered every 15 minutes (using Sql_trigger_value or datagroup_trigger).
Our looker regenerator is set for default value.( checks every 5 minutes)
Issue: A new version of all the PDTs (with 15 mins as trigger value) are created by regenerator every 15 minutes. Maximum up to 6 versions. After that, all the duplicates dropped and only one version exist and the cycle continues.(meaning after fifteen minutes, a duplicate is created and so on up to 6 versions).
I have set the persist_for to “0 minutes” at explore level and as well as view level.
Also I have set max_cache_age to “0 minutes”.
I just couldn't understand why looker takes 90 minutes to drop the duplicates and why is this not the case in production?
My database is Amazon Redshift and My looker Version is 7.10.