Find the answer to your Linux question:
Results 1 to 5 of 5
This is my first attempt to install Linux (let alone Gentoo). I'm trying to install Gentoo on an HP Proliant ML350 G4 with a single Intel Xeon processor and Smart ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. #1
    Just Joined!
    Join Date
    Jun 2005
    Posts
    1

    Block device /dev/cciss/c0d0p3 is not a valid root device...


    This is my first attempt to install Linux (let alone Gentoo). I'm trying to install Gentoo on an HP Proliant ML350 G4 with a single Intel Xeon processor and Smart Array 641 (3 SCSI drives in RAID 5). I'm following the handbook for x86.

    I've made two attempts through the complete install, with many boot attempts each time while tweaking various settings based upon discussion of similar problems on this and other forums. The reason for the second attempt was to try changing the filesystem of my root partition from ext3 to reiserfs (to no avail). I've tried both the GRUB and LILO boot loaders with the same result. When I attempt to boot from the hard drive array, the boot goes smoothly until this point:

    Code:
    >> Mounting filesystems
    >> Activating udev...
    >> Determining root device...
    >> Block device /dev/cciss/c0d0p3 is not a valid root device...
    >> The root block device is unspecified or not detected.
       Please specify a device to boot, or “shell” for a shell...
    I partitioned my hard drive array as follows:

    Code:
    # fdisk /dev/cciss/c0d0
    
    Command (m for help): p
    
    Disk /dev/cciss/c0d0: 145.6 GB, 145661483520 bytes
    255 heads, 63 sectors/track, 17709 cylinders
    Units = cylinders of 16065 * 512 = 8225280 bytes
    
               Device Boot   Start      End      Blocks    Id   System
    /dev/cciss/c0d0p1   *        1        5       40131    83   Linux
    /dev/cciss/c0d0p2            6       68      506047+   82   Linux swap / Solaris
    /dev/cciss/c0d0p3           69     1936    15004710    83   Linux
    /dev/cciss/c0d0p4         1937    17709   126696622+   83   Linux
    I installed stage tarball stage3-pentium4-2005.0.tar.bz2.

    I chose to configure the 2.6 kernel and I used genkernel.

    Presently, I'm using LILO. Here is my /etc/lilo.conf:

    Code:
    # more /etc/lilo.conf
    boot=/dev/cciss/c0d0
    prompt
    timeout=50
    default=gentoo
    
    image=/boot/kernel-2.6.11-gentoo-r11
      label=gentoo
      read-only
      root=/dev/ram0
      append="init=/linuxrc ramdisk=8192 real_root=/dev/cciss/c0d0p3 udev doscsi"
      initrd=/boot/initrd-2.6.11-gentoo-r11
    Previously, I was using GRUB. Here is my /boot/grub/grub.conf:

    Code:
    # more /boot/grub/grub.conf
    default 0
    timeout 30
    splashimage=(hd0,0)/grub/splash.xpm.gz
    
    title=Gentoo Linux 2.6.11-r11
    root (hd0,0)
    kernel /kernel-2.6.11-gentoo-r11 root=/dev/ram0 init=/linuxrc ramdisk=8192 real_root=/dev/cciss/c0d0p3 udev doscsi
    initrd /initrd-2.6.11-gentoo-r11
    Here is my /etc/fstab:

    Code:
    # more /etc/fstab
    /dev/cciss/c0d0p1      /boot           ext2            defaults,noatime        1 2
    /dev/cciss/c0d0p3      /               reiserfs        noatime                 0 1
    /dev/cciss/c0d0p2      none            swap            sw                      0 0
    /dev/cciss/c0d0p4      /home           ext3            noatime                 0 0
    /dev/cdroms/cdrom0     /mnt/cdrom      iso9660         noauto,ro               0 0
    /dev/fd0               /mnt/floppy     auto            noauto                  0 0
    none                   /proc           proc            defaults                0 0
    none                   /dev/shm        tmpfs           nodev,nosuid,noexec     0 0
    Here is my /etc/make.conf (I made no changes):

    Code:
    # more /etc/make.conf
    CFLAGS="-02 -march=pentium4 -fomit-frame-pointer"
    CHOST="i686-pc-linux-gnu"
    CXXFLAGS="${CFLAGS}"

    As I said above, I receive the same error regardless of using GRUB or LILO, so the problem seems to be bigger than the configuration of the boot loader.

    Any help or ideas would be greatly appreciated. I've run out of new things to try.

  2. #2
    Linux Guru loft306's Avatar
    Join Date
    Oct 2003
    Location
    The DairyLand
    Posts
    1,666
    hmmm im in the same boat.....though i remember booting raid 5 once on proliants i can seem to get it to go $%^&* some ppl say raid5 is un bootable but i have seen thease boot raid5

    tnhough u did get it a bit farther than i did.....off to play with mine...i'll see if i came up witha an ansawer
    just keep on playing with lilo i dont think grub will boot it at all atleast not in raid5 and you /we may need to specify a geometry address to boot from(the hardware address for that hdd/partition)

    but im just still geussing here myself..... the laost install that i got raid5 to go on a proliant was with devfs though so i just think were playing a naming game here

    oh you have the
    cciss in the kernel or loading as a module at boot?
    ~Mike ~~~ Forum Rules
    Testing? What's that? If it compiles, it is good, if it boots up, it is perfect. ~ Linus Torvalds
    http://loft306.org

  3. #3
    Linux Guru
    Join Date
    Oct 2001
    Location
    Täby, Sweden
    Posts
    7,578
    I don't know or anything, but to make an educated guess: The kernel obviously detects the correct device, since it can mount and boot from the root filesystem. Then the startup scripts fail to detect it. My conclusion is that udev is probably naming it differently.

    If you choose to go to a shell when it says so, what do you see in /dev? Is /dev/cciss there?

  4. $spacer_open
    $spacer_close
  5. #4
    Linux Guru loft306's Avatar
    Join Date
    Oct 2003
    Location
    The DairyLand
    Posts
    1,666
    Quote Originally Posted by Dolda2000
    I don't know or anything, but to make an educated guess: The kernel obviously detects the correct device, since it can mount and boot from the root filesystem. Then the startup scripts fail to detect it. My conclusion is that udev is probably naming it differently.

    If you choose to go to a shell when it says so, what do you see in /dev? Is /dev/cciss there?
    yeah and no....but probably correct though in udev it seems to stay the same and it was in devfs that the name used to change...but who knows
    hey try this for realrooton the kernel line in grub or lilo
    Code:
    root=/dev/ida/c0d1p1 or realroot
    though it might also be root=/dev/cciss0/c0d0p1
    that is how udev sees my scsi arrays
    whereis the /dev/cciss/c0d0p1 is needed to get it to the point that it is atto use the controller driver or is it tto get the array seen for boot (i know im explaining badly but im still a bit confused with booting raid5 myself)
    what i glean is this..the cciss driver is needed to get it to boot but the udev device name is /dev/ida/c0d1p1 and should be in fstab or maybe not in fstab play with it and let me know if it worked .......i think me just figged it out...im off to try it on my box

    and as you have probably geussed at this point it is realy just getting teh bootloader rite and the problem goes no further than that
    ~Mike ~~~ Forum Rules
    Testing? What's that? If it compiles, it is good, if it boots up, it is perfect. ~ Linus Torvalds
    http://loft306.org

  6. #5
    Just Joined!
    Join Date
    Nov 2008
    Posts
    1

    cciss boot

    I had the same problem. I use a gentoo, and I used genkernel to configure the kernel. I solved the problem using /dev/cciss!c0d0p3 in the grub root configuration.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •