Configuring RAID 10 with RedHat ES 4 (64 bit)

From: Andy Mountford (andrew.mountford_at_gmail.com)
Date: 05/18/05

  • Next message: Burke, Thomas G.: "RE: MBR" 
    Date: Wed, 18 May 2005 10:57:49 +0100
To: redhat-list@redhat.com

    Hi,

    I am trying to configure an array of 12 disks in a RAID 10
    configuration. All 12 disks are identical 146Gb drives.

    First off, on each disk I created a single partition which spanned the
    whole disk with a system id of 0xFD. After writing the partition
    table, a print confirmed all looked good.

    Secondly, I used mdadm to create 6 RAID 1 pairs. (mirrored).

    mdadm --create /dev/md0 --level=1 -n 2 /dev/sda /dev/sdb
    mdadm --create /dev/md1 --level=1 -n 2 /dev/sdc /dev/sdd
    mdadm --create /dev/md2 --level=1 -n 2 /dev/sde /dev/sdf
    mdadm --create /dev/md3 --level=1 -n 2 /dev/sdg /dev/sdh
    mdadm --create /dev/md4 --level=1 -n 2 /dev/sdi /dev/sdj
    mdadm --create /dev/md5 --level=1 -n 2 /dev/sdk /dev/sdl

    cat /proc/mdstat showed the devices syncing. Once this had completed I
    created a final meta device which striped the RAID1 meta devices. e.g.

    mdadm --create /dev/md6 --level 0 -n 6 /dev/md0 dev/md1 dev/md2
    dev/md3 dev/md4 dev/md5

    I could then create a filesystem on the new array and mount it - all
    looked good. However, 2 things concerned me:

    1. After creating the RAID array, the partition table for all the
    disks involved was empty and an fdisk returned:

    "Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)"

    What's going on here?

    2. How can I get the array to come up after a reboot? What needs to be
    done? Do I need an rc script that runs mdadm?

    Thanks,

    Andy 

    -- 
redhat-list mailing list
unsubscribe mailto:redhat-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list
  • Next message: Burke, Thomas G.: "RE: MBR"

    Relevant Pages

    • Re: NT4 upgrade to W2K3 with new hardware
      ... I'd rather think it's some kind of RAID. ... it's much better to divide disks rather than make only one big array and then partition it. ... If you create one big array and partition it, you're doing I/O on all the disks for every disk operation, regardless of the partition involved, so you're degrading performance. ...
      (microsoft.public.windows.server.active_directory)
    • Re: Making a RAID array
      ... hard it might be to set up my own RAID array. ... and then set it up to do RAID. ... disks to that, but I'd like to see how easy it is to do it in the Mac ... If I add disks to the mirror set, ...
      (uk.comp.sys.mac)
    • Re: Debian+LVM+RAID
      ... When the swap partition quits working, ... good hardware RAID controller and likely put the swap partition onto ... Aren't you making it hard by having to partition all the disks ... I had the disks partitioned and had created RAID arrays ...
      (Debian-User)
    • Re: insufficient space in super block for rotational layout
      ... Occurs typically on very high density disks. ... the file system structure cannot ... Is this really an issue with a RAID 5 storage ... It doesn't sound like a problem with the array, but the size of the array. ...
      (comp.unix.solaris)
    • Re: VLDB, RAID, and clustering
      ... And clustering for windows and SQL Server is a hardware failover technology only, ... You can not have data or log files on disks outside the cluster resource groups. ... You absolutely need to ensure the transaction logs for all user and tempdb databases are separate from the data files and on a raid 1 or 10. ... You may need to segregate the tempdb data files onto their own array as well depending on usage. ...
      (microsoft.public.sqlserver.security)