When enabling TLS connections for Postgres as documented here Configuring PostgreSQL TLS, the DB container logs:
LOG: could not receive data from client: Connection reset by peer
There’s no impact about this log, however, you may want to understand why this is repeatedly logged.
This is an expected normal behavior. The reason why it happens is because Kong maintains a connection pool with Postgres, and the default timeout is 60 seconds, which is controlled by the pg_keepalive_timeout parameter. When a connection in the pool times out, it will be released. If TLS is enabled on that connection, the underlying Nginx that Kong depends on will send an RST packet to quickly close TLS when closing the connection. The RST packet caused Postgres to detect that Kong reset the connection and therefore output a log.