On Fri, Jan 20, 2017 at 12:08 AM, Karl O. Pinc <k...@meme.com> wrote: > Is this an argument for having the current_logfiles always exist > and be empty when there is no in-filesystem logfile? It always felt > to me that the code would be simpler that way.
Well, you'll need to do something in any case if the logging_collector is found disabled and the syslogger process is restarted. So just removing it looked cleaner to me. I am not strongly attached to one way of doing or the other though. -- Michael -- Sent via pgsql-hackers mailing list (email@example.com) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers