Re: initdb and fsync

Jeff Davis <pgsql@j-davis.com>

From: Jeff Davis <pgsql@j-davis.com>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Noah Misch <noah@leadboat.com>, Tom Lane <tgl@sss.pgh.pa.us>, Andrew Dunstan <andrew@dunslane.net>, pgsql-hackers@postgresql.org
Date: 2012-06-23T09:17:45Z
Lists: pgsql-hackers
On Fri, 2012-06-22 at 10:04 -0400, Robert Haas wrote:
> This may be a stupid question, by why is it initdb's job to fsync the
> files the server creates, rather than the server's job?  Normally we
> rely on the server to make its own writes persistent.

That was my first reaction as well:

http://archives.postgresql.org/message-id/1328468000.15224.32.camel@jdavis

However, from the discussion it seems like it will be harder to do it
that way (during normal operation, a checkpoint is what syncs the files;
but as Tom points out, bootstrap mode does not). Also, I would expect
the fsyncs to be in a less-controlled pattern, so it might perform more
poorly.

Regards,
	Jeff Davis