Results 40 comments of abc3

@RishabhPachori, the application name property `supavisor_meta` indicates that these connections are used by Supavisor as a metadata store, where it keeps data about tenants. [By default, the metadata connection pool...

@RishabhPachori, let me explain further. Supavisor is a multitenant solution, which means it needs to store data about different tenants, databases, users, and more. For this purpose, it uses a...

@RishabhPachori, LGTM. Those connections without `application_name` are likely related to the analytics service. But since I'm no longer part of the Supabase team, I can only help you with Supavisor

> leaking connections In case of failure in the following cases, the node will not handle notifications because there is no retry logic. So yeah, there’s a potential place for...

@hauleth ~fyi https://github.com/supabase/libcluster_postgres/pull/24~ I've closed that PR because there's already one in progress that adds auto_reconnect. So, to fix the current issue, just need to add auto_reconnect to the notification...

> Supavisor is only used if you are using for your own DB connections. It will come in as the role you use. @GaryAustin1, supavisor also uses the customer's database...

If you want to fix #616 by switching to libcluster_postgres, it won’t help because it’s just a copy of supavisor's module and has the same potential buggy behavior.

Out of curiosity, did you try using Supavisor on port 5432? Does the issue still persist?

@istarkov could you reach out to Supabase support to check your project's logs?