Provider slow to refresh schema resources

[Intuinewin]

Hello,

It has become very complicated to apply our terraform project which contains schema resources.

After investigations, the query SELECT 1 FROM SVL_QUERY_SUMMARY takes several minutes to run on our cluster (1m41 last time I ran it) and we declare a lot of schemas in our tf (~80). I'm wondering if another view could be used to do the multi-az check

Thanks for your help

Related: #12

[mmichaelb]

Hi! I get that running this command in a cluster slows down things but to this day, I have not found a different command/approach that checks where the provider is running. We do already cache response but with these long response times, that does not help a lot I guess. Maybe we can find other ways to determine the Redshift type or we can also think about adding a provider config param to give users the option to specify the type ahead of the terraform/tofu lifecycle.

[Intuinewin]

I think new configuration parameters for serverless and multi-AZ Redshift would be a good idea. This would simplify things and avoid magic queries to detect the cluster type.