we cache connections based on the log name, but presumably someone could issue a query to force the cache and the log table to not be in sync. this would only happen by explicitly re-configuring the metadata without restarting backends, but it would be nice to be able to handle it some how.
we cache connections based on the log name, but presumably someone could issue a query to force the cache and the log table to not be in sync. this would only happen by explicitly re-configuring the metadata without restarting backends, but it would be nice to be able to handle it some how.