Date: Tue, 22 Jan 2019 10:17:02 -0800
From: Andres Freund <andres@anarazel.de>
To: Kevin Grittner <kgrittn@gmail.com>
Cc: Thomas Munro <thomas.munro@enterprisedb.com>,
	Dmitry Dolgov <9erthalion6@gmail.com>,
	Robert Haas <robertmhaas@gmail.com>,
	PostgreSQL-development <pgsql-hackers@postgresql.org>,
	Shawn Debnath <sdn@amazon.com>
Subject: Re: Refactoring the checkpointer's fsync request queue
Message-ID: <20190122181702.jueacsyffjwla3bb@alap3.anarazel.de>
References: 
 <CA+TgmoZRNHDKSnG35mPufKKgVXSPAYrQo5nHhD_ZzBzg50cTAQ@mail.gmail.com>
 <CAEepm=3c6ybA7zUCQEuFees7o6JGhNGdS3V7ay3EdCU6avDaaA@mail.gmail.com>
 <CA+TgmobJDfSZ2x_vic5ZoheVcJBn-AmnndUs_99vzP9VStnBuA@mail.gmail.com>
 <CAEepm=10A5eySmc9=sPje6nNbaSBYB0gQ3Rb0dDvph-Ei2H0cQ@mail.gmail.com>
 <CAEepm=1TDor1xHkYtpqEpuu5kpoU-DbLGX49scCY_zY-t3H-ug@mail.gmail.com>
 <CA+q6zcV=p=1G2Z8itJ9=tAn=b8ha3pO4RQCMt9YXOEutRV33YQ@mail.gmail.com>
 <CAEepm=2qBbQXGeekhCY+W8m=fy-+eU_FAUzRCSPFKuZ2rKG1LQ@mail.gmail.com>
 <CAEepm=1uqZLcTqLz_+egZRYotnTvzDK-jOGZdN8QhL=7nixY5Q@mail.gmail.com>
 <CAEepm=1mp+7PPY+sKJ4cvs6hba0NnWeed00oJ=EWry0gUgnJ_A@mail.gmail.com>
 <CACjxUsNAtKw-ARB=OEPde5j3e18Xza5xsBBsBrZxJw5nK1LLgQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CACjxUsNAtKw-ARB=OEPde5j3e18Xza5xsBBsBrZxJw5nK1LLgQ@mail.gmail.com>
Precedence: bulk

Hi,,

On 2019-01-22 08:27:48 -0600, Kevin Grittner wrote:
> With the help of VMware's Dirk Hohndel (VMware's Chief Open Source
> Officer, a VP position near the top of the organization, and a
> personal friend of Linus), I have been fortunate enough to make
> contact directly with Linus Torvalds to discuss this issue.  In emails
> to me he has told me that this patch is no longer provisional:
> 
>   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=fff75eb2a08c2ac96404a2d79685668f3cf5a7a3

Unfortunately, unless something has changed recently, that patch is
*not* sufficient to really solve the issue - we don't guarantee that
there's always an fd preventing the necessary information from being
evicted from memory:

 Note that we might still lose the error if the inode gets evicted from
 the cache before anything can reopen it, but that was the case before
 errseq_t was merged. At LSF/MM we had some discussion about keeping
 inodes with unreported writeback errors around in the cache for longer
 (possibly indefinitely), but that's really a separate problem"

And that's entirely possibly in postgres.  The commit was dicussed on
list too, btw...

Greetings,

Andres Freund