All of lore.kernel.org
 help / color / mirror / Atom feed
From: Benjamin Marzinski <bmarzins@redhat.com>
To: device-mapper development <dm-devel@redhat.com>
Subject: Re: Configuring multipath for root device
Date: Fri, 21 Dec 2007 10:26:09 -0600	[thread overview]
Message-ID: <20071221162609.GL25939@ether.msp.redhat.com> (raw)
In-Reply-To: <110b17190712200846l50e4c4d8h51456b852f0c851d@mail.gmail.com>

On Thu, Dec 20, 2007 at 08:46:41AM -0800, Pradipmaya Maharana wrote:
> Hi,
> 
> Pardon my knowledge on Linux, but in your article did you mean that
> the /boot should be on a local disk and /root, swap can be on
> multipath device.
> 
> I am trying to install /boot, /root and swap all on one multipath
> device and try to boot from it. It does not boot, no error just blank
> screen. My equirement is to install everything on multipath device; so
> is it possible at all? I mean can I have the boot device (/boot) too
> on a multipath device.
> 
> Following is what I tried on RHEL
>  - selected "linux mpath"
>  - I could see one local disk and 3 multipath devices
>  - I chose custom partition type
>  - instead of going for LVM, I selected one single multipath device
> and installed everything on it (/boot, / and swap).
>  - Tried rebooting, it won't boot :(
> 
> What am I missing here?

I haven't ever tried putting swap on a multipath device, but I believe
it should work.  Your steps seem correct. Are you getting any messages
at all on the console when you boot up?

One thing you need to be aware of is that while multipathed root is
pretty simple, for multipathed boot to work, your machine needs to be
able to boot from the device that your are multipathing with no help
from any linux device drivers (since they obviously can't be loaded
until you run the boot loader).  I believe some machines need something
to be set in BIOS in order to try and boot off SAN storage.

To narrow down the problem, I would try setting up the same
configuration, but without multipathing the device. If this doesn't
work, then you are probably having problems booting from your storage.
If it does work, then when you set it up with multipathing, you should
at least be making it past GRUB, and looking at the console output
should shed some light on what's going wrong.

Note, if you have a device that needs a prio_callout, multipath will
still fail in the all paths down case, because it can't access the
callout program.  The upstream code currently fixes that by using
callout library functions, and the next device-mapper-multipath package
from RHEL will fix it by caching the callouts in a private namespace.

-Ben

> Thanks and Regards,
> Pradipmaya.
> 
> 
> On 12/19/07, Pradipmaya Maharana <pradipmaya@gmail.com> wrote:
> > Hi Chandra,
> >
> > Thanks. I will givei it a try.
> >
> > Regards,
> > Pradipmaya.
> >
> > On 12/19/07, Chandra Seetharaman <sekharan@us.ibm.com> wrote:
> > > I am working on a multipath usage document. Attached is the installation
> > > section of that document.
> > >
> > > Note that it is written for RHEL5 U1 and SLES10 SP1.
> > >
> > > Good Luck,
> > >
> > > chandra
> > >
> > > On Wed, 2007-12-19 at 13:02 -0800, Pradipmaya Maharana wrote:
> > > > Hi All,
> > > >
> > > > Is there a process/document that talks about how to configure
> > > > root/boot device for multipath?
> > > >
> > > > Thanks and Regards,
> > > > Pradipmaya.
> > > >
> > > > --
> > > > dm-devel mailing list
> > > > dm-devel@redhat.com
> > > > https://www.redhat.com/mailman/listinfo/dm-devel
> > > --
> > >
> > > ----------------------------------------------------------------------
> > >    Chandra Seetharaman               | Be careful what you choose....
> > >              - sekharan@us.ibm.com   |      .......you may get it.
> > > ----------------------------------------------------------------------
> > >
> > >
> > > --
> > > dm-devel mailing list
> > > dm-devel@redhat.com
> > > https://www.redhat.com/mailman/listinfo/dm-devel
> > >
> > >
> >
> 
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

  reply	other threads:[~2007-12-21 16:26 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-12-19 21:02 Configuring multipath for root device Pradipmaya Maharana
2007-12-19 21:18 ` Chandra Seetharaman
2007-12-19 21:26   ` Pradipmaya Maharana
2007-12-20 16:46     ` Pradipmaya Maharana
2007-12-21 16:26       ` Benjamin Marzinski [this message]
2007-12-21 19:06         ` Pradipmaya Maharana
2008-01-07  7:38   ` Hannes Reinecke
2008-01-07 18:11     ` Pradipmaya Maharana
2008-01-07 20:19   ` Pradipmaya Maharana
2008-01-07 20:28     ` malahal
2008-01-07 20:40       ` Pradipmaya Maharana
2008-01-07 20:46         ` malahal
2008-01-07 20:56           ` Pradipmaya Maharana
2008-01-07 21:57             ` malahal
2008-01-08  0:13               ` Pradipmaya Maharana
2008-01-08  0:29                 ` malahal
2008-01-08  7:17     ` Hannes Reinecke
2008-01-16  2:25       ` Pradipmaya Maharana
2008-01-16  2:32         ` Chandra Seetharaman
2008-01-16 17:36         ` malahal
2008-01-16 18:49           ` Pradipmaya Maharana
2008-01-17 20:21             ` malahal
2008-01-17 20:38               ` Pradipmaya Maharana
2008-01-18 21:52                 ` Pradipmaya Maharana
2008-01-16  4:24       ` Romanowski, John (OFT)

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=20071221162609.GL25939@ether.msp.redhat.com \
    --to=bmarzins@redhat.com \
    --cc=dm-devel@redhat.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 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.