The current PostgreSQL adapter checks if the schema exists whenever the tenant is changed. Our application connects to a new tenant on every request. While the schema check is usually fast, with a high request throughput it starts to add up. Summed up per day, our database spends a lot of time verifying the same thing over and over again.
We have a fixed list of available tenants. We need to restart anyway when we add a new tenant. At the point where Apartment.tenant_names includes a tenant, we are certain that the schema exists. And if not, it's fine to blow up.
It would be great if there was a possibility to disable this check with a configuration of sorts.
The current PostgreSQL adapter checks if the schema exists whenever the tenant is changed. Our application connects to a new tenant on every request. While the schema check is usually fast, with a high request throughput it starts to add up. Summed up per day, our database spends a lot of time verifying the same thing over and over again.
We have a fixed list of available tenants. We need to restart anyway when we add a new tenant. At the point where
Apartment.tenant_namesincludes a tenant, we are certain that the schema exists. And if not, it's fine to blow up.It would be great if there was a possibility to disable this check with a configuration of sorts.