From: Bill Davidsen <davidsen@tmr.com>
To: Bryan Mesich <bryan.mesich@ndsu.edu>, Neil Brown <neilb@suse.de>,
Jon@eHardcastle.com, linux-raid@vger.kernel.org
Subject: Re: Why does one get mismatches?
Date: Tue, 16 Feb 2010 16:25:25 -0500 [thread overview]
Message-ID: <4B7B0D45.7040801@tmr.com> (raw)
In-Reply-To: <20100211175133.GA30187@atlantis.cc.ndsu.nodak.edu>
Bryan Mesich wrote:
> On Thu, Feb 11, 2010 at 04:14:44PM +1100, Neil Brown wrote:
>
>>> This whole discussion simply shows that for RAID-1 software RAID is less
>>> reliable than hardware RAID (no, I don't mean fake-RAID), because it
>>> doesn't pin the data buffer until all copies are written.
>>>
>> That doesn't make it less reliable. It just makes it more confusing.
>>
>
> I agree that linux software RAID is no less reliable than
> hardware RAID with regards to the above conversation. It's
> however confusing to have a counter that indicates there are
> problems with a RAID 1 array when in fact there is not.
>
Sorry, but real hardware raid is more reliable than software raid, and
Neil's justification for not doing smart recovery mentions it. Note this
referes to real hardware raid, not fakeraid which is just some firmware
in a BIOS to use the existing hardware.
The issue lies with data changing between write to multiple drives. In
hardware raid the data traverses the memory bus once, only once, and
goes into cache in the controller, from which it is written to all
mirrored drives. With software raid an individual write is done to each
drive, and if the data in the buffer changes between writes to one drive
or the other you get different values. Neil may be convinced that the OS
somehow "knows" which of the mirror copies is correct, ie. most recent,
and never uses the stale data, but if that information was really
available reads would always return the latest value and it wouldn't be
possible to read the same file multiple times and get different MD5sums.
It would also be possible to do a stable smart recovery by propagating
the most recent copy to the other mirror drives.
I hoped that mounting data=journal would lead to consistency, that seems
not to be true either.
--
Bill Davidsen <davidsen@tmr.com>
"We can't solve today's problems by using the same thinking we
used in creating them." - Einstein
next prev parent reply other threads:[~2010-02-16 21:25 UTC|newest]
Thread overview: 104+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-20 11:52 Fw: Why does one get mismatches? Jon Hardcastle
2010-01-22 18:13 ` Goswin von Brederlow
2010-01-24 17:40 ` Jon Hardcastle
2010-01-24 21:52 ` Roger Heflin
2010-01-24 23:13 ` Goswin von Brederlow
2010-01-25 10:07 ` Jon Hardcastle
2010-01-25 10:37 ` Goswin von Brederlow
2010-01-25 10:52 ` Jon Hardcastle
2010-01-25 17:32 ` Goswin von Brederlow
2010-01-25 19:32 ` Iustin Pop
2010-02-01 21:18 ` Bill Davidsen
2010-02-01 22:37 ` Neil Brown
2010-02-02 15:11 ` Bill Davidsen
2010-02-03 11:17 ` Goswin von Brederlow
2010-02-11 5:14 ` Neil Brown
2010-02-11 17:51 ` Bryan Mesich
2010-02-16 21:25 ` Bill Davidsen [this message]
2010-02-16 21:38 ` Steven Haigh
2010-02-17 3:19 ` Bryan Mesich
2010-02-17 23:05 ` Neil Brown
2010-02-19 15:18 ` Piergiorgio Sartor
2010-02-19 22:02 ` Neil Brown
2010-02-19 22:37 ` Piergiorgio Sartor
2010-02-19 23:34 ` Asdo
2010-02-20 4:27 ` Goswin von Brederlow
2010-02-20 11:12 ` Asdo
2010-02-21 11:13 ` Goswin von Brederlow
[not found] ` <8754A21825504719B463AD9809E54349@m5>
[not found] ` <20100221194400.GA2570@lazy.lzy>
2010-02-22 13:01 ` Asdo
2010-02-22 13:30 ` Piergiorgio Sartor
2010-02-22 13:44 ` Piergiorgio Sartor
2010-02-24 19:42 ` Bill Davidsen
2010-02-20 4:23 ` Goswin von Brederlow
2010-02-24 14:54 ` Bill Davidsen
2010-02-24 21:37 ` Neil Brown
2010-02-26 20:48 ` Bill Davidsen
2010-02-26 21:09 ` Neil Brown
2010-02-26 22:01 ` Piergiorgio Sartor
2010-02-26 22:15 ` Bill Davidsen
2010-02-26 22:21 ` Piergiorgio Sartor
2010-02-26 22:20 ` Asdo
2010-02-27 6:01 ` Michael Evans
2010-02-28 0:01 ` Bill Davidsen
2010-02-24 14:46 ` Bill Davidsen
2010-02-24 16:12 ` Martin K. Petersen
2010-02-24 18:51 ` Piergiorgio Sartor
2010-02-24 22:21 ` Neil Brown
2010-02-25 8:41 ` Piergiorgio Sartor
2010-03-02 4:57 ` Neil Brown
2010-03-02 18:49 ` Piergiorgio Sartor
2010-02-24 21:39 ` Neil Brown
[not found] ` <4B8640A2.4060307@shiftmail.org>
2010-02-25 10:41 ` Neil Brown
2010-02-28 8:09 ` Luca Berra
2010-03-02 5:01 ` Neil Brown
2010-03-02 7:36 ` Luca Berra
2010-03-02 10:04 ` Michael Evans
2010-03-02 11:02 ` Luca Berra
2010-03-02 12:13 ` Michael Evans
2010-03-02 18:14 ` Asdo
2010-03-02 18:52 ` Piergiorgio Sartor
2010-03-02 23:27 ` Asdo
2010-03-03 9:13 ` Piergiorgio Sartor
2010-03-03 11:42 ` Asdo
2010-03-03 12:03 ` Piergiorgio Sartor
2010-03-02 20:17 ` Neil Brown
2010-02-24 21:32 ` Neil Brown
2010-02-25 7:22 ` Goswin von Brederlow
2010-02-25 7:39 ` Neil Brown
2010-02-25 8:47 ` John Robinson
2010-02-25 9:07 ` Neil Brown
2010-02-11 18:12 ` Piergiorgio Sartor
-- strict thread matches above, loose matches on Subject: below --
2010-02-01 23:14 Jon Hardcastle
2010-01-25 20:43 greg
2010-01-25 22:49 ` Steven Haigh
2010-01-27 21:54 ` Tirumala Reddy Marri
2010-01-28 9:16 ` Jon Hardcastle
2010-01-28 10:29 ` Asdo
2010-01-28 17:20 ` Tirumala Reddy Marri
2010-01-28 18:23 ` Goswin von Brederlow
2010-01-28 19:03 ` Tirumala Reddy Marri
2010-01-28 20:24 ` Goswin von Brederlow
2010-01-29 15:37 ` Jon Hardcastle
2010-01-29 23:52 ` Goswin von Brederlow
2010-01-30 10:39 ` Jon Hardcastle
2010-02-01 21:10 ` Bill Davidsen
2010-01-20 15:03 Jon Hardcastle
2010-01-20 15:34 ` Brett Russ
2010-01-20 20:44 ` Majed B.
2010-01-20 22:25 ` Brett Russ
2010-01-20 22:30 ` Majed B.
2010-01-20 22:43 ` Brett Russ
2010-01-20 23:01 ` Christopher Chen
2010-01-21 4:17 ` Steven Haigh
2010-01-21 8:08 ` Asdo
2010-01-21 10:52 ` Steven Haigh
2010-01-21 11:48 ` Farkas Levente
2010-01-21 12:15 ` Jon Hardcastle
2010-01-19 10:04 Jon Hardcastle
2010-01-20 14:19 ` Brett Russ
2010-01-20 14:34 ` Jon Hardcastle
2010-01-20 14:46 ` Brett Russ
2010-02-01 20:48 ` Bill Davidsen
2010-01-22 16:22 ` Jon Hardcastle
2010-01-22 16:34 ` Asdo
2010-01-22 17:41 ` Brett Russ
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=4B7B0D45.7040801@tmr.com \
--to=davidsen@tmr.com \
--cc=Jon@eHardcastle.com \
--cc=bryan.mesich@ndsu.edu \
--cc=linux-raid@vger.kernel.org \
--cc=neilb@suse.de \
/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).