linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Stark <gsstark@mit.edu>
To: Jeff Garzik <jgarzik@pobox.com>
Cc: Greg Stark <gsstark@mit.edu>, Mike Fedyk <mfedyk@matchmail.com>,
	Erik Steffl <steffl@bigfoot.com>,
	linux-kernel@vger.kernel.org
Subject: Re: libata in 2.4.24?
Date: 02 Dec 2003 15:10:19 -0500	[thread overview]
Message-ID: <877k1f9e1g.fsf@stark.dyndns.tv> (raw)
In-Reply-To: <20031202190646.GA9043@gtf.org>

Jeff Garzik <jgarzik@pobox.com> writes:

> So, today, no acknowledgement occurs until the data _really_ is in the
> drive's buffers.

The drive's buffers isn't good enough. If power is lost the write will be lost
and the database corrupt. It needs to be on the platters.


> > This doesn't happen with SCSI disks where multiple requests can be pending so
> > there's no urgency to reporting a false success. The request doesn't complete
> > until the write hits disk. As a result SCSI disks are reliable for database
> > operation and IDE disks aren't unless write caching is disabled.
> 
> This is not really true.
> 
> Regardless of TCQ, if the OS driver has not issued a FLUSH CACHE (IDE)
> or SYNCHRONIZE CACHE (SCSI), then the data is not guaranteed to be on
> the disk media.  Plain and simple.

That doesn't agree with people's experience. People seem to find that SCSI
drives never cache writes. This sort of makes sense since there's just not
much reason to report a write success before the write can be performed.
There's no performance advantage as long as more requests can be queued up.


> If fsync(2) returns without a flush-cache, then your data is not
> guaranteed to be on the disk.  And as you noted, flush-cache destroys
> performance.

It's my understanding that it doesn't. There was some discussion in the past
month about making the drivers issue syncs for journalled filesystems, but
even then the idea of adding it to fsync or O_SYNC files wasn't the
motivation.


> There are three levels:
> 
> a) Data is successfully transferred to the controller/drive queue (TCQ).
> b) Data is successfully transferred to the drive's internal buffers.
> c) The drive successfully transfers data to the media.

Only the third is of interest to Postgres or other databases. In fact, I
suspect only the third is of interest to other systems that are supposed to be
reliable like MTAs etc. I think Wietse and others would be shocked if they
were told fsync wasn't guaranteed to have waited until the writes had actually
hit the media.

-- 
greg


  reply	other threads:[~2003-12-02 20:10 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-11-28 18:27 linux-2.4.23 released Marcelo Tosatti
2003-11-28 19:06 ` Willy Tarreau
2003-11-28 22:55 ` J.A. Magallon
2003-11-29 22:26 ` libata in 2.4.24? Samuel Flory
2003-11-29 23:10   ` Marcelo Tosatti
2003-12-01 10:43     ` Marcelo Tosatti
2003-12-01 18:06       ` Samuel Flory
2003-12-01 21:12         ` Greg Stark
2003-12-01 21:23           ` Samuel Flory
2003-12-01 21:44             ` Greg Stark
2003-12-01 22:00               ` Jeff Garzik
2003-12-01 22:06               ` Samuel Flory
2003-12-01 22:00             ` Erik Steffl
2003-12-02  5:36               ` Greg Stark
     [not found]                 ` <20031202055336.GO1566@mis-mike-wstn.matchmail.com>
2003-12-02  5:58                   ` Mike Fedyk
2003-12-02 16:31                     ` Greg Stark
2003-12-02 17:40                       ` Mike Fedyk
2003-12-02 18:04                         ` Jeff Garzik
2003-12-02 18:46                           ` Mike Fedyk
2003-12-02 18:49                             ` Jeff Garzik
2003-12-04  8:18                         ` Jens Axboe
2003-12-02 18:02                       ` Jeff Garzik
2003-12-02 18:51                         ` Greg Stark
2003-12-02 19:06                           ` Jeff Garzik
2003-12-02 20:10                             ` Greg Stark [this message]
2003-12-02 20:16                               ` Jeff Garzik
2003-12-02 20:34                                 ` Greg Stark
2003-12-02 22:34                               ` bill davidsen
2003-12-02 23:02                                 ` Mike Fedyk
2003-12-02 23:18                                   ` bill davidsen
2003-12-02 23:40                                     ` Mike Fedyk
2003-12-03  0:01                                     ` Jeff Garzik
2003-12-03  0:47                                 ` Jamie Lokier
2003-12-07  5:33                                   ` Bill Davidsen
2003-12-01 21:36           ` Justin Cormack
2003-12-01 13:41 Xose Vazquez Perez
2003-12-01 14:11 ` Marcelo Tosatti
2003-12-02 19:59   ` Stephan von Krawczynski
2003-12-02 22:05   ` bill davidsen
2003-12-02 22:34     ` Jeff Garzik
2003-12-03  0:34 Xose Vazquez Perez

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=877k1f9e1g.fsf@stark.dyndns.tv \
    --to=gsstark@mit.edu \
    --cc=jgarzik@pobox.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mfedyk@matchmail.com \
    --cc=steffl@bigfoot.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).