From: Stephen Satchell <satch@concentric.net>
To: "H. Peter Anvin" <hpa@zytor.com>, linux-kernel@vger.kernel.org
Subject: Re: Journaling pointless with today's hard disks?
Date: Sat, 24 Nov 2001 15:23:30 -0800 [thread overview]
Message-ID: <4.3.2.7.2.20011124150445.00bd4240@10.1.1.42> (raw)
In-Reply-To: <9tp6tg$mge$1@cesium.transmeta.com>
In-Reply-To: <tgpu68gw34.fsf@mercury.rus.uni-stuttgart.de> <20011124103642.A32278@vega.ipal.net> <20011124184119.C12133@emma1.emma.line.org> <tgy9kwf02c.fsf@mercury.rus.uni-stuttgart.de>
At 02:28 PM 11/24/01 -0800, H. Peter Anvin wrote:
> > > However, if it's really true that DTLA drives and their successor
> > > corrupt blocks (generate bad blocks) on power loss during block writes,
> > > these drives are crap.
> >
> > They do, even IBM admits that (on
> >
> > http://www.cooling-solutions.de/dtla-faq
> >
> > you find a quote from IBM confirming this). IBM says it's okay, you
> > have to expect this to happen. So much for their expertise in making
> > hard disks. This makes me feel rather dizzy (lots of IBM drives in
> > use).
> >
>
>No sh*t. I have always been favouring IBM drives, and I had a RAID
>system with these drives bought. It will be a LONG time before I buy
>another IBM drive, that's for sure. I can't believe they don't even
>have the decency of saying "we fucked".
It is the responsibility of the power monitor to detect a power-fail event
and tell the drive(s) that a power-fail event is occurring. If power goes
out of specification before the drive completes a commanded write, what do
you expect the poor drive to do? ANY glitch in the write current will
corrupt the current block no matter what -- the final CRC isn't
recorded. Most drives do have a panic-stop mode when they detect voltage
going out of range so as to minimize the damage caused by an
out-of-specification power-down event, and more importantly use the energy
in the spinning platter to get the heads moved to a safe place before the
drive completely spins down. The panic-stop mode is EXACTLY like a Linux
OOPS -- it's a catastrophic event that SHOULD NOT OCCUR.
Most power supplies are not designed to hold up for more than 30-60 ms at
full load upon removal of mains power. Power-fail detect typically
requires 12 ms (three-quarters cycle average at 60 Hz) or 15 ms
(three-quarters cycle average at 50 Hz) to detect that mains power has
failed, leaving your system a very short time to abort that long queue of
disk write commands. It's very possible that by the time the system wakes
up to the fact that its electron feeding tube is empty it has already
started a write operation that cannot be completed before power goes out of
specification. It's a race condition.
Fix your system.
If you don't have a UPS on that RAID, and some means of shutting down the
RAID gracefully when mains power goes, you are sealing your own doom,
regardless of the maker of the hard drive you use in that RAID. Even the
original CDC disk drives, some of the best damn drives ever manufactured in
the world, would corrupt data when power failed during a write.
Satch
next prev parent reply other threads:[~2001-11-24 23:24 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <mailman.1006644421.6553.linux-kernel2news@redhat.com>
2001-11-24 13:03 ` Florian Weimer
2001-11-24 13:40 ` Rik van Riel
2001-11-24 16:36 ` Phil Howard
2001-11-24 17:19 ` Charles Marslett
2001-11-24 17:31 ` Florian Weimer
2001-11-24 17:41 ` Matthias Andree
2001-11-24 19:20 ` Florian Weimer
2001-11-24 19:29 ` Rik van Riel
2001-11-24 22:51 ` John Alvord
2001-11-24 23:41 ` Phil Howard
2001-11-25 0:24 ` Ian Stirling
2001-11-25 0:53 ` Phil Howard
2001-11-25 1:25 ` H. Peter Anvin
2001-11-25 1:44 ` Sven.Riedel
2001-11-24 22:28 ` H. Peter Anvin
2001-11-25 4:49 ` Andre Hedrick
2001-11-24 23:04 ` Pedro M. Rodrigues
2001-11-24 23:23 ` Stephen Satchell [this message]
2001-11-24 23:29 ` H. Peter Anvin
2001-11-26 18:05 ` Steve Brueggeman
2001-11-26 23:49 ` Martin Eriksson
2001-11-27 0:06 ` Andreas Dilger
2001-11-27 0:16 ` Andre Hedrick
2001-11-27 7:38 ` Andreas Dilger
2001-11-27 11:48 ` Ville Herva
2001-11-27 0:18 ` Jonathan Lundell
2001-11-27 1:01 ` Ian Stirling
2001-11-27 1:33 ` H. Peter Anvin
2001-11-27 1:57 ` Steve Underwood
2001-11-27 5:04 ` Stephen Satchell
2001-11-25 4:20 ` Pete Zaitcev
2001-11-25 12:30 ` Matthias Andree
2001-11-25 15:04 ` Barry K. Nathan
2001-11-25 16:31 ` Matthias Andree
2001-11-27 2:39 ` Pavel Machek
2001-12-03 10:23 ` Matthias Andree
2001-11-25 9:14 ` Chris Wedgwood
2001-11-25 22:55 ` Daniel Phillips
2001-11-26 16:59 ` Rob Landley
2001-11-26 20:30 ` Andre Hedrick
2001-11-26 20:35 ` Rob Landley
2001-11-26 23:59 ` Andreas Dilger
2001-11-27 0:24 ` H. Peter Anvin
2001-11-27 0:52 ` H. Peter Anvin
2001-11-27 1:11 ` Andrew Morton
2001-11-27 1:15 ` H. Peter Anvin
2001-11-27 16:59 ` Matthias Andree
2001-11-27 16:56 ` Matthias Andree
2001-11-27 1:23 ` Ian Stirling
2001-11-26 23:00 ` Rob Landley
2001-11-27 2:41 ` H. Peter Anvin
2001-11-27 0:19 ` Rob Landley
2001-11-27 23:35 ` Andreas Bombe
2001-11-28 14:32 ` Rob Landley
2001-11-27 3:39 ` Ian Stirling
2001-11-27 7:03 ` Ville Herva
2001-11-27 16:50 ` Matthias Andree
2001-11-27 20:31 ` Rob Landley
2001-11-28 18:43 ` Matthias Andree
2001-11-28 18:46 ` Rob Landley
2001-11-28 22:19 ` Matthias Andree
2001-11-29 22:21 ` Pavel Machek
2001-12-01 10:55 ` Jeff V. Merkey
2001-12-02 0:08 ` Matthias Andree
2001-12-03 20:04 ` Pavel Machek
2001-11-26 20:53 ` Richard B. Johnson
2001-11-26 21:18 ` Journaling pointless with today's hard disks? [wandering OT] Rob Landley
2001-11-27 0:32 ` Journaling pointless with today's hard disks? H. Peter Anvin
2001-11-27 16:39 ` Matthias Andree
2001-11-27 17:42 ` Martin Eriksson
2001-11-28 16:35 ` Ian Stirling
2001-11-26 17:14 ` Steve Brueggeman
2001-11-26 20:36 ` Andre Hedrick
2001-11-26 21:14 ` Steve Brueggeman
2001-11-26 21:36 ` Andre Hedrick
2001-11-27 16:36 ` Steve Brueggeman
2001-11-27 20:04 ` Bill Davidsen
2001-11-27 21:28 ` Wayne Whitney
2001-11-27 21:52 ` Andre Hedrick
2001-11-28 11:53 ` Pedro M. Rodrigues
2001-11-25 13:52 ` Pedro M. Rodrigues
2001-11-25 1:20 dnu478nt5w@mailexpire.com
2001-11-28 14:36 Galappatti, Kishantha
2001-11-28 17:22 David Balazic
2001-11-28 23:25 Frank de Lange
2001-11-29 1:52 ` Matthias Andree
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=4.3.2.7.2.20011124150445.00bd4240@10.1.1.42 \
--to=satch@concentric.net \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
/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
all inboxes | Powered by JetHome®