From: Neil Brown <neilb@suse.de>
To: martin f krafft <madduck@madduck.net>
Cc: Piergiorgio Sartor <piergiorgio.sartor@nexgo.de>,
567468@bugs.debian.org, Daniel Reurich <daniel@centurion.net.nz>,
linux-raid <linux-raid@vger.kernel.org>
Subject: Re: md homehost (was: Bug#567468: (boot time consequences of) Linux mdadm superblock) question.
Date: Tue, 23 Feb 2010 13:30:28 +1100 [thread overview]
Message-ID: <20100223133028.279e0174@notabene.brown> (raw)
In-Reply-To: <20100222091632.GE2641@lapse.rw.madduck.net>
On Mon, 22 Feb 2010 10:16:32 +0100
martin f krafft <madduck@madduck.net> wrote:
> also sprach Piergiorgio Sartor <piergiorgio.sartor@nexgo.de> [2010.02.21.2113 +0100]:
> > I do not see how the "homehost" plays a role, here.
>
> Neil,
>
> Could you please put forth the argument for why the homehost must
> match, and why unconditional auto-assembly is not desirable?
> Realistically, what problems are we protecting against?
>
The problem to protect against is any consequence of rearranging devices
while the host is off, including attaching devices that previously were
attached to a different computer.
mdadm will currently assembly any array that it finds, but will not give a
predictable name to anything that looks like it might be imported from a
different host.
So if you have 'md0' on each of two computers, one computer dies and you move
the devices from that computer to the other, then as long as the bios boots
of the right drive, mdadm will assemble the local array as 'md0' and the
other array as 'something else'.
There are two ways that mdadm determines than array is 'local'.
1/ is the uuid listed against an array in mdadm.conf
2/ is the 'homehost' encoded in the metadata.
If either of those is true, the array is local and gets a predictable name.
If neither, the name gets an _%d suffix.
This is only an issue if you use device name (.e.g /dev/md0 or /dev/md/root)
to mount the root filesystem.
If you use mount-by-uuid then it clearly doesn't matter what name mdadm
assembles the array under. In that case, the fs UUID (stored on the
initramfs or similar) will assure the necessary uniqueness and mdadm need not
worry about homehost.
But if '/' is mounted by a name in /dev/md/, I want to be sure mdadm puts the
correct array at that name no matter what other arrays might be visible.
Does that clarify things enough?
NeilBrown
next prev parent reply other threads:[~2010-02-23 2:30 UTC|newest]
Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-14 1:51 Linux mdadm superblock question Volker Armin Hemmann
2010-02-14 4:02 ` Michael Evans
2010-02-14 7:21 ` david
2010-02-14 8:38 ` Michael Evans
2010-02-14 18:40 ` Volker Armin Hemmann
2010-02-14 18:53 ` John Robinson
2010-02-14 21:16 ` Gabor Gombas
[not found] ` <201002142013.24922.volkerarmin@googlemail.com>
2010-02-16 14:28 ` John Robinson
2010-02-16 14:37 ` Volker Armin Hemmann
2010-02-16 14:46 ` Robin Hill
2010-02-16 17:23 ` John Robinson
2010-02-16 19:38 ` Luca Berra
2010-02-16 17:18 ` Bill Davidsen
2010-02-16 21:06 ` Volker Armin Hemmann
2010-02-16 22:00 ` Nick Bowler
2010-02-16 22:18 ` Volker Armin Hemmann
2010-02-17 14:25 ` Nick Bowler
2010-02-18 9:27 ` Ian Dall
2010-02-17 1:03 ` Mr. James W. Laferriere
2010-02-17 2:01 ` Neil Brown
2010-02-17 2:38 ` Volker Armin Hemmann
2010-02-17 23:15 ` Neil Brown
2010-02-17 6:34 ` Kyle Moffett
2010-02-17 9:38 ` Rudy Zijlstra
2010-02-17 13:26 ` Frans Pop
2010-02-17 20:54 ` Gabor Gombas
2010-02-17 21:29 ` Frans Pop
2010-02-18 3:40 ` Goswin von Brederlow
2010-02-17 16:22 ` Kyle Moffett
2010-02-17 17:41 ` david
2010-02-17 18:10 ` Nick Bowler
2010-02-17 18:27 ` Volker Armin Hemmann
2010-02-17 18:37 ` Nick Bowler
2010-02-17 18:41 ` david
2010-02-17 18:51 ` Nick Bowler
2010-02-17 21:17 ` david
2010-02-17 21:37 ` Nick Bowler
2010-02-17 22:21 ` david
2010-02-17 22:29 ` boot times, not mdadm (was: Linux mdadm superblock question.) martin f krafft
2010-02-17 23:24 ` (boot time consequences of) Linux mdadm superblock question Neil Brown
2010-02-17 23:50 ` martin f krafft
2010-02-18 2:58 ` Henrique de Moraes Holschuh
2010-02-18 3:26 ` martin f krafft
2010-02-18 4:03 ` Daniel Reurich
2010-02-18 4:40 ` martin f krafft
2010-02-18 5:10 ` Neil Brown
2010-02-18 5:21 ` martin f krafft
2010-02-18 5:34 ` Neil Brown
2010-02-19 0:42 ` martin f krafft
2010-02-19 2:51 ` Daniel Reurich
[not found] ` <20100221171445.GB17267@lapse.rw.madduck.net>
2010-02-22 7:06 ` Goswin von Brederlow
2010-02-22 7:37 ` Bug#567468: " Michael Evans
2010-02-22 9:14 ` martin f krafft
2010-02-22 9:11 ` martin f krafft
2010-02-22 10:42 ` Daniel Reurich
2010-02-19 9:16 ` Piergiorgio Sartor
[not found] ` <20100221174007.GB19058@lapse.rw.madduck.net>
[not found] ` <20100221201304.GB2570@lazy.lzy>
2010-02-22 9:16 ` Bug#567468: md homehost (was: Bug#567468: (boot time consequences of) Linux mdadm superblock) question martin f krafft
2010-02-22 11:11 ` Daniel Reurich
2010-02-23 7:29 ` md homehost Goswin von Brederlow
2010-02-23 8:10 ` martin f krafft
2010-02-23 2:30 ` Neil Brown [this message]
2010-02-23 6:27 ` md homehost (was: Bug#567468: (boot time consequences of) Linux mdadm superblock) question martin f krafft
2010-02-23 7:31 ` md homehost Goswin von Brederlow
2010-02-23 8:16 ` Bug#567468: " martin f krafft
2010-02-24 13:13 ` Goswin von Brederlow
2010-02-24 17:52 ` Mario 'BitKoenig' Holbe
2010-02-24 22:23 ` Neil Brown
2010-02-23 8:18 ` Piergiorgio Sartor
2010-02-24 0:10 ` md homehost (was: Bug#567468: (boot time consequences of) Linux mdadm superblock) question Neil Brown
2010-02-24 4:12 ` Michael Evans
2010-02-24 13:41 ` md homehost Goswin von Brederlow
2010-02-24 22:30 ` Neil Brown
2010-02-25 7:16 ` Goswin von Brederlow
2010-02-25 7:46 ` Neil Brown
2010-02-25 8:33 ` Michael Evans
2010-02-25 11:55 ` Mario 'BitKoenig' Holbe
2010-02-18 5:17 ` (boot time consequences of) Linux mdadm superblock question Daniel Reurich
2010-02-18 5:22 ` martin f krafft
2010-02-17 18:46 ` Volker Armin Hemmann
2010-02-17 22:26 ` H. Peter Anvin
2010-02-18 3:33 ` Goswin von Brederlow
2010-02-18 7:51 ` Luca Berra
2010-02-18 14:12 ` Nick Bowler
2010-02-19 9:04 ` Michael Evans
2010-02-14 19:34 ` Henrique de Moraes Holschuh
2010-02-14 20:07 ` Michael Evans
2010-02-14 21:14 ` Henrique de Moraes Holschuh
2010-02-14 20:47 ` Asdo
2010-02-14 21:26 ` Henrique de Moraes Holschuh
2010-02-14 21:28 ` Gabor Gombas
2010-02-15 9:08 ` martin f krafft
2010-02-15 7:51 ` Luca Berra
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=20100223133028.279e0174@notabene.brown \
--to=neilb@suse.de \
--cc=567468@bugs.debian.org \
--cc=daniel@centurion.net.nz \
--cc=linux-raid@vger.kernel.org \
--cc=madduck@madduck.net \
--cc=piergiorgio.sartor@nexgo.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).