by-passing RAID controller

[Logu_2]

I have a BL360 G5 server. It has a h/w raid controller. Entering the RAID controller bios shows 3 options (raid 0, raid 1+0 and raid 5). I dont want any raid at all at this time. How do I configure the server not to use the RAID controller.

Currently it is configure for RAID 0. There are 3 disks on the bay. First disk has the OS and two other disks were added later.

Can I remove the other two disks or the data could have spread across all the disks since it is configured for RAID 0.

[TTr]

You can not bypass the raid controller and use the disks directly by the OS.

If your current raid0 contains all 3 disks, yes the data is spread across all 3 disks, "it is striped", so you can not remove the 2 disks without destroying the 3-disk raid0 and creating one or more 1-disk raid0.

It looks you want individual disks in your OS level, but why? What are you after? Maybe it is doable with HW raid.

[Logu_2]

It doesn't look like using the other two disks. So I just want to make use of other two disks to backup a few things.

[root@www ~]# sfdisk -s
/dev/cciss/c0d0: 71652960
/dev/cciss/c0d1: 71652960
/dev/cciss/c0d2: 71652960
/dev/dm-0: 10485760
/dev/dm-1: 10485760
/dev/dm-2: 10485760
/dev/dm-3: 4194304
/dev/dm-4: 10485760
/dev/dm-5: 20971520
total: 282067744 blocks
[root@www ~]# cat /proc/partitions
major minor #blocks name

104 0 71652960 cciss/c0d0
104 1 204800 cciss/c0d0p1
104 2 71444480 cciss/c0d0p2
104 16 71652960 cciss/c0d1
104 32 71652960 cciss/c0d2
253 0 10485760 dm-0
253 1 10485760 dm-1
253 2 10485760 dm-2
253 3 4194304 dm-3
253 4 10485760 dm-4
253 5 20971520 dm-5
[root@www ~]# ll /dev/block/104:2
lrwxrwxrwx 1 root root 15 2010-05-10 00:25 /dev/block/104:2 -> ../cciss/c0d0p2
[root@www ~]# mount
/dev/mapper/vg_visolve-lv_root on / type ext4 (rw)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
devpts on /dev/pts type devpts (rw,gid=5,mode=620)
tmpfs on /dev/shm type tmpfs (rw)
/dev/cciss/c0d0p1 on /boot type ext4 (rw)
/dev/mapper/vg_visolve-lv_home on /home type ext4 (rw)
/dev/mapper/vg_visolve-lv_usr on /usr type ext4 (rw)
/dev/mapper/vg_visolve-lv_var on /var type ext4 (rw)
/dev/mapper/vg_visolve-lv_data on /DATA type ext4 (rw)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)

[TTr]

I don't recognize the OS but iit smells like Solaris (x86). I can not tell from the OS commands if the other two disks are used or not. You need to check it at the array level.
How many raid-0 arrays do you have? one or three?
If you have one raid0 array you need to check how many disks it has in it. See if you can check this in the raid controller bios when the server boots up.

I am guessing you want to use zfs for disk/filesystem right?

ACU is not available for Solaris if that's your OS. You might even lose some performance because of lack of drivers for the raid controller in this case.
http://h18000.www1.hp.com/products/servers/proliantstorage/software-management/acumatrix/index.html

[Logu_2]

This is linux (Fedora). I dont have the console access at the moment. I will try the array config on the bios when I get a chance.

There are 3 disks on the server.
First disk is partitioned into two. One for /boot and the other is sole member of the volume group which is then divided into logical volumes for /usr /var swap etc.

Now I want to use other 2 disks for backup. But the raid-0 thing is confusing a bit

[TTr]

You should install ACU for Linux from the URL above. Also take a look at the raid technology so that you understand the concept of raid0 and other raid types.
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c00687518/c00687518.pdf

[Pieter 't Hart]

You allways traverse the controller.

I think you could just create an raid-0 array with only one disk.