From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
Cc: pgsql-performance@postgresql.org, Josh Berkus <josh@agliodbs.com>
Cc: Greg Smith <greg@2ndquadrant.com>
Subject: Re: Defaulting wal_sync_method to fdatasync on Linux for 9.1?
Date: Tue, 16 Nov 2010 20:17:43 -0500
Message-ID: <11392.1289956663@sss.pgh.pa.us> (raw)
In-Reply-To: <11113.1289955868@sss.pgh.pa.us>
References: <AANLkTi=n3zxiLNFhbpiN=JPQPwB=BhXf9iepw1KOEs8U@mail.gmail.com>
<201011170130.10720.andres@anarazel.de>
<10822.1289955088@sss.pgh.pa.us>
<201011170201.24782.andres@anarazel.de>
<11113.1289955868@sss.pgh.pa.us>
I wrote:
> The fsync is associated with the write, which is not done with insert
> lock held. We're not quite that dumb.
But wait --- are you thinking of the call path where a write (and
possible fsync) is forced during AdvanceXLInsertBuffer because there's
no WAL buffer space left? If so, that's *exactly* the scenario that
can be expected to be less common with more buffer space.
regards, tom lane
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: tgl@sss.pgh.pa.us, andres@anarazel.de, josh@agliodbs.com, greg@2ndquadrant.com
Subject: Re: Defaulting wal_sync_method to fdatasync on Linux for 9.1?
In-Reply-To: <11392.1289956663@sss.pgh.pa.us>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox