From: Phil Turmel <philip@turmel.org>
To: Stefan Lamby <webmaster@peter-speer.de>,
"linux-raid@vger.kernel.org" <linux-raid@vger.kernel.org>
Subject: Re: Raid 10 Issue - Booting in case raid failed
Date: Fri, 06 Mar 2015 15:12:37 -0500 [thread overview]
Message-ID: <54FA0A35.8040304@turmel.org> (raw)
In-Reply-To: <322760937.22899.1425668800663.JavaMail.open-xchange@app04.ox.hosteurope.de>
On 03/06/2015 02:06 PM, Stefan Lamby wrote:
> Hi list.
>
> If everything will work out OK, I will end up with an raid 10 array with 4
> devices.
>
> My partition design and layout structure will be found at the end, if needed.
>
> There are a few questions left for me in case I have to boot with a failed disk:
>
> 1) As you might have seen from the partition design, only partition sda1 has the
> boot flag set. As far as I guess, the ubuntu installer was using grub-install
> only for sda. I am kind of afraid what will happen, in case sda will fail in the
> future. Will it be a good idea to grub-install to all the other devices also?
Yes, sort of.
> 2) What about the boot flag, if I need to grub-install the other devices also?
> Should it be O or 1? Do I have to leave it set to false and in case things go
> wrong boot from a live cd and set it to on to boot from another device?
Set to true.
But the "sort of" comes from your reliance on grub support for MD raid,
and having the appropriate mirrors containing the boot folder. You've
left enough space before your first partition (48g) to easily hold a
plain raid1 x4 boot partition and a raid6 root partition (use a small
chunk size for that). Then your system could boot with any two drives
missing, and let you know what's possible with the large raid10.
Also note that this kind of boot redundancy only helps if the bad drive
is entirely missing at boot time. If you really need boot redundancy,
you have fewer choices: BIOS fakeraid, or hardware raid with a BIOS
extension, or EFI boot with a monolithic kernel/initramfs on each device.
HTH,
Phil
next prev parent reply other threads:[~2015-03-06 20:12 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-05 17:56 Raid 10 Issue Stefan Lamby
2015-03-05 20:07 ` Phil Turmel
2015-03-06 10:09 ` Raid 10 Issue - Swapping Data from Array to Array Stefan Lamby
2015-03-06 12:57 ` Phil Turmel
2015-03-09 8:51 ` Raid 10 Issue - Swapping Data from Array to Array [SOLVED] Stefan Lamby
2015-03-06 19:06 ` Raid 10 Issue - Booting in case raid failed Stefan Lamby
2015-03-06 20:12 ` Phil Turmel [this message]
2015-03-08 14:59 ` Raid 10 Issue Wilson, Jonathan
2015-03-06 8:54 ` Robin Hill
2015-03-06 9:32 ` Raid 10 Issue [SOLVED] Stefan Lamby
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=54FA0A35.8040304@turmel.org \
--to=philip@turmel.org \
--cc=linux-raid@vger.kernel.org \
--cc=webmaster@peter-speer.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.