We have configured "pglogical" in AWS Aurora. But the "subscription" process keeps going down frequently. We don't get enough log in database error log to know exactly what's causing the problem (we have tried setting the "log_min_messages" and "pglogical.conflict_log_level" upto "debug5") . From the replication slot, we can get the SQL statement where apply process breaks, but running that statement manually on the target side works just fine. Replication process is configured with "postgres" (admin user). Is there a way we can force the "apply" process to spill more info in the error log file? Or some other way to troubleshoot?