From: John Bradford <john@grabjohn.com>
To: skraw@ithnet.com (Stephan von Krawczynski)
Cc: john@grabjohn.com (John Bradford),
josh@stack.nl, alan@lxorguk.ukuu.org.uk,
linux-kernel@vger.kernel.org
Subject: Re: Are linux-fs's drive-fault-tolerant by concept?
Date: Mon, 21 Apr 2003 10:55:12 +0100 (BST) [thread overview]
Message-ID: <200304210955.h3L9tCVx000292@81-2-122-30.bradfords.org.uk> (raw)
In-Reply-To: <20030421113236.3955d5e6.skraw@ithnet.com> from "Stephan von Krawczynski" at Apr 21, 2003 11:32:36 AM
> > > > > > Fault tolerance in a filesystem layer means in practical terms
> > > > > > that you are guessing what a filesystem should look like, for the
> > > > > > disk doesn't answer that question anymore. IMHO you don't want
> > > > > > that to be done automagically, for it might go right sometimes,
> > > > > > but also might trash everything on RW filesystems.
> > > > >
> > > > > Let me clarify again: I don't want fancy stuff inside the filesystem
> > > > > that magically knows something about right-or-wrong. The only _very
> > > > > small_ enhancement I would like to see is: driver tells fs there is an
> > > > > error while writing a certain block => fs tries writing the same
> > > > > data onto another block. That's it, no magic, no RAID
> > > > > stuff. Very simple.
> > > >
> > > > That doesn't belong in the filesystem.
> > > >
> > > > Imagine you have ten blocks free, and you allocate data to all of them
> > > > in the filesystem. The write goes to cache, and succeeds.
> > > >
> > > > 30 seconds later, the write cache is flushed, and an error is reported
> > > > back from the device.
> > >
> > > And where's the problem?
> > > Your case:
> > > Immediate failure. Disk error.
> > >
> > > My case:
> > > Immediate failure. Disk error (no space left for replacement)
> > >
> > > There's no difference.
> >
> > In my case, the machine can continue as normal. The filesystem is
> > intact, (with no blocks free). The block device driver has to cope
> > with the error, which could be as simple as holding the data in RAM
> > until an operator has been paged to replace the disk.
>
> Forgive my ignorance, but I have not seen a case up to today where
> ide, aicX or 3ware has called me up for a replacement unit, written
> to it and been ok afterwards. What the heck are you talking of?
Modern disks error correct on write. The user doesn't even know that
they are doing it. If a disk actually reports back a write error, it
is usually very broken.
Incidentally, I don't see how your idea would even be implementable
without disabling write caching.
> I am not really interested in what a low-level driver could do
> unless there is none that does it...
I assume you mean 'one that does it'.
If nobody was interested in what a low-level driver could do unless
there was one that does it already, we wouldn't be innovating anything
new.
> And again, how do you think this should work out on your _root_
> partition? (see below)
1. Hot plug a new disk
2. Duplicate the read-only root file system on to it
3. Pivot root
> > In your case, the filesystem is no longer in a usable state.
>
> I have yet to see an fs thats in a writeable state after the medium
> is full ...
It is perfectly writable, for example, for a delete operation.
> > If that
> > was the root filesystem, the machine will, at best, probably go in to
> > single user mode, with a read-only root filesystem.
>
> How come?
In my opinion, that would be the best course of action, if the device
holding the root filesystem is faulty.
> > > Thing is: If there are 11 blocks free and not ten, then you fail
> >
> > Wrong. See above.
>
> Please tell me when you were last "paged to replace the disk"? If you can't
> tell me, then you know I am right by now.
I have never been paged to replace a disk by a Linux system.
That is why I would like to see this functionality added to Linux.
John.
next prev parent reply other threads:[~2003-04-21 9:40 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-04-19 16:04 Are linux-fs's drive-fault-tolerant by concept? Stephan von Krawczynski
2003-04-19 15:29 ` Alan Cox
2003-04-19 17:00 ` Stephan von Krawczynski
2003-04-19 22:04 ` Alan Cox
2003-04-20 16:24 ` Stephan von Krawczynski
2003-04-20 13:59 ` John Bradford
2003-04-20 16:55 ` Stephan von Krawczynski
2003-04-20 17:12 ` John Bradford
2003-04-20 17:21 ` Stephan von Krawczynski
2003-04-20 18:48 ` Alan Cox
2003-04-20 20:00 ` John Bradford
2003-04-21 1:51 ` jw schultz
2003-04-19 21:13 ` Jos Hulzink
2003-04-20 16:07 ` Stephan von Krawczynski
2003-04-20 16:40 ` John Bradford
2003-04-20 17:01 ` Stephan von Krawczynski
2003-04-20 17:20 ` John Bradford
2003-04-21 9:32 ` Stephan von Krawczynski
2003-04-21 9:55 ` John Bradford [this message]
2003-04-21 11:24 ` Stephan von Krawczynski
2003-04-21 11:50 ` Alan Cox
2003-04-21 12:14 ` John Bradford
2003-04-19 16:22 ` John Bradford
2003-04-19 16:36 ` Russell King
2003-04-19 16:45 ` John Bradford
2003-04-19 16:52 ` Stephan von Krawczynski
2003-04-19 20:04 ` John Bradford
2003-04-19 20:33 ` Andreas Dilger
2003-04-21 9:25 ` Denis Vlasenko
2003-04-21 9:42 ` John Bradford
2003-04-21 10:25 ` Stephan von Krawczynski
2003-04-21 10:50 ` John Bradford
2003-04-19 20:38 ` Stephan von Krawczynski
2003-04-20 14:21 ` John Bradford
2003-04-21 9:09 ` Denis Vlasenko
2003-04-21 9:35 ` John Bradford
2003-04-21 11:03 ` Stephan von Krawczynski
2003-04-21 12:04 ` John Bradford
2003-04-21 11:22 ` Denis Vlasenko
2003-04-21 11:46 ` Stephan von Krawczynski
2003-04-21 12:13 ` John Bradford
2003-04-19 20:05 ` John Bradford
2003-04-19 23:13 ` Arnaldo Carvalho de Melo
2003-04-19 17:54 ` Felipe Alfaro Solana
2003-04-25 0:07 ` Stewart Smith
2003-04-25 0:52 ` Richard B. Johnson
2003-04-25 7:13 ` John Bradford
[not found] ` <20030419161011$0136@gated-at.bofh.it>
2003-04-19 17:18 ` Florian Weimer
2003-04-19 18:07 ` Stephan von Krawczynski
2003-04-19 18:41 ` Dr. David Alan Gilbert
2003-04-19 20:56 ` Helge Hafting
2003-04-19 21:15 ` Valdis.Kletnieks
2003-04-20 10:51 ` Helge Hafting
2003-04-20 19:04 ` Valdis.Kletnieks
2003-04-19 21:57 ` Alan Cox
2003-04-20 10:09 ` Geert Uytterhoeven
2003-04-21 8:37 ` Denis Vlasenko
2003-05-05 12:38 ` Pavel Machek
2003-04-19 22:02 ` Alan Cox
2003-04-20 8:41 ` Arjan van de Ven
2003-04-25 0:11 ` Stewart Smith
2003-04-20 15:06 Chuck Ebbert
2003-04-20 15:19 ` John Bradford
2003-04-20 17:03 Chuck Ebbert
2003-04-20 17:25 ` John Bradford
2003-04-20 17:28 Chuck Ebbert
2003-04-21 9:36 ` Stephan von Krawczynski
2003-04-20 17:28 Chuck Ebbert
2003-04-20 17:44 Chuck Ebbert
2003-04-20 17:44 Chuck Ebbert
[not found] <mail.linux.kernel/20030420185512.763df745.skraw@ithnet.com>
[not found] ` <03Apr21.020150edt.41463@gpu.utcc.utoronto.ca>
2003-04-21 11:19 ` Stephan von Krawczynski
2003-04-21 11:52 ` Alan Cox
2003-04-21 14:14 ` Valdis.Kletnieks
2003-05-06 7:03 ` Mike Fedyk
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=200304210955.h3L9tCVx000292@81-2-122-30.bradfords.org.uk \
--to=john@grabjohn.com \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=josh@stack.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=skraw@ithnet.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).