Re: Help installing grub

[Yan Seiner]

Jordan Uggla wrote:
> On Tue, May 29, 2012 at 7:30 PM, Yan Seiner <address@hidden> wrote:
>   
> > I need some help with a custom install.
> >
> > I have a client that I access via a remote connection.  Their server has two
> > physical drives set up in a software RAID-1.  To make a long story short,
> > through a "perfect storm" series of events, they lost both hard drives.  I
> > shipped them a backup of their data and their local admin got the server
> > running on a single drive.
> >
> > I figured I could set up the other drive as 1/2 of a RAID array, install the
> > data, and then reboot the server with the RAID running and complete the
> > mirroring operation.  I mostly followed this process:
> >  https://wiki.archlinux.org/index.php/Convert_a_single_drive_system_to_RAID
> >
> > I am running into trouble getting a valid boot sector on the new raid drive.
> >  This is the first time I've dealt with GRUB 2, and I am not sure how to set
> > it up to boot off of /dev/md1 (the boot partition).
> >
> > Here's what I have:
> >
> > /dev/sdb - current boot drive with GRUB installed correctly.  This boots
> > just fine.
> > /dev/sda - new half-array drive without GRUB installed.
> >
> > Q:  How do I get GRUB installed so it boots off of /dev/md1 (/dev/sda2)?
> >     
>
> To install grub you need to use grub-install. If you were already in
> the booted system, with the array already setup normally, you would
> simply run "grub-install /dev/sda && grub-install /dev/sdb". It
> doesn't sound like that's the case for you, but it's hard to tell and
> that information is critical. How / where is /dev/md1 going to be
> mounted when you run grub-install?
>   
The system is currently booted off /dev/sdb1.  sdb has a different 
partitioning scheme.  It's my goal to reboot the system with the raid 
running, then repartition /dev/sdb to add the missing half of the raid 
array.

The relevant parts of the system setup on the disk that I want to boot from:

/etc/fstab:
/dev/md0    /        ext4    errors=remount-ro 0    0
/dev/md1    /boot        ext4    defaults    0    1

/proc/mdstat:
md1 : active (auto-read-only) raid1 sda2[1]
     248896 blocks [2/1] [_U]
md0 : active raid1 sda1[1]
     979840 blocks [2/1] [_U]

fdisk /dev/sda -l
  Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1         122      979933+  fd  Linux raid 
autodetect
/dev/sda2   *         123         153      249007+  fd  Linux raid 
autodetect



>   
> > I am doing this all via ssh; I have no physical access to the server, so
> > some way of checking that the system will actually boot correctly would be
> > welcome.
> >     
>
> This is possible via qemu (but you need to be *very* careful). I will
> go into more detail on this after you've answered my above question.
>
>   


--
Help us raise money for my kids' swim team!

http://www.facebook.com/RiverRoadKids4Kids