Horton

Re: fcntl(SETLK) [was Re: 2nd update on TOAST]

Bruce Momjian <pgman@candle.pha.pa.us>

View thread
From: Bruce Momjian <pgman@candle.pha.pa.us>
To: Peter Eisentraut <peter_e@gmx.net>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Jan Wieck <JanWieck@Yahoo.com>, Philip Warner <pjw@rhyme.com.au>, PostgreSQL HACKERS <pgsql-hackers@postgresql.org>
Date: 2000-07-08T15:03:07Z
Lists: pgsql-hackers 
> The other problem is a socket file left behind by a crashed postmaster. I
> don't consider this such a big problem; a crashed postmaster is not the
> normal mode of operation. The friendly message we have right now seems
> alright to me. And it's a way of tell that the postmaster crashed at all.
> 
> One idea to get the pid in there somewhere is creating a socket file
> "/tmp/.s.PGSQL.port.pid" and making /tmp/.s.PGSQL.port a symlink to it.
> Then clients don't know the difference, but the server knows the pid and
> can take appropriate action. Or make the symlink the other way around, not
> sure.

The symlink is an interesting idea.  lstat() on the normal name can give
the file name with pid.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026