Member Avatar for Casperjames 
           Device Boot      Start         End      Blocks   Id  System
/dev/cciss/c0d0p1   *        2048      499711      248832   83  Linux
/dev/cciss/c0d0p2          501758   573366271   286432257    5  Extended
/dev/cciss/c0d0p5          501760   573366271   286432256   83  Linux

Okay, we are building a new data server that has 6 disks - we partitioned the first 4 as a RAID 1 and the last 2 as a RAID 0. Our RAID 1 will house the main data server and the RAID 0 we want to use as a NAS/file storage device. My question is which partition is the RAID 0? Is the Extended partition the RAID 0? I want to mount 0 on the 1 as a NAS but I don't want to write the files to the wrong device.

Thanks

Edited by Casperjames
  • 2 Contributors
  • 1 Reply
  • 637 Views
  • 9 Hours Discussion Span
  • Latest Post Latest Post by rch1231

Dani AI

Generated

Short answer: the Extended partition you posted is just part of the partition table on the first virtual disk the controller exposed — it is not the RAID‑0 array itself. If you created a RAID‑0 on the last two physical disks via the hardware controller, the controller will present that RAID as a separate virtual/logical disk (a separate block device) to Linux. The cciss driver exposes controller logical volumes and keeps controller info under /proc/driver/cciss, so check there or with the controller utility to see every virtual disk the card created. (manpages.ubuntu.com)

To identify which /dev is the RAID‑0: use the RAID card’s management CLI or the BIOS/UEFI config utility. For HPE SmartArray use the SSA/ssacli tool to list controllers, logical drives and the physical members; for LSI/Avago/Broadcom cards use storcli/MegaCli. Example checks (run as root or sudo) are shown below — the exact path/tool depends on your vendor:

sudo ssacli ctrl all show config
sudo /opt/MegaRAID/storcli/storcli64 /c0 show all

Those commands will show which physical bays are in each logical drive so you can match the RAID‑0 to the device the OS should see. (ctera.com)

Once the controller has exposed the RAID‑0, identify the OS device and bind it stably before mounting. Use lsblk / blkid and the udev symlinks in /dev/disk/by-id or mount by UUID in /etc/fstab so you do not accidentally mount the wrong disk. Example:

lsblk -o NAME,SIZE,FSTYPE,UUID,MOUNTPOINT
ls -l /dev/disk/by-id/
# then add to /etc/fstab using UUID=...

Using UUID or by-id is far more reliable than raw /dev names. (help.ubuntu.com)

A final caution: RAID‑0 gives NO redundancy — a single disk failure wipes the whole array. If that storage will hold anything important, keep reliable backups or consider RAID levels with parity/mirroring (RAID‑5/RAID‑1/RAID‑6) as suggested. Also note that older systems used the cciss driver but newer kernels moved to hpsa and device naming/paths can change across kernel versions — verify the controller utility output whenever you create or change arrays. (dell.com)

Member Avatar for rch1231

Hello,

What you are seeing here are two partitions both on the first controller and the first disk.
The device naming scheme is:

       /dev/cciss/c0d0         Controller 0, disk 0, whole device
       /dev/cciss/c0d0p1       Controller 0, disk 0, partition 1
       /dev/cciss/c0d0p2       Controller 0, disk 0, partition 2
       /dev/cciss/c0d0p3       Controller 0, disk 0, partition 3

       /dev/cciss/c1d1         Controller 1, disk 1, whole device
       /dev/cciss/c1d1p1       Controller 1, disk 1, partition 1
       /dev/cciss/c1d1p2       Controller 1, disk 1, partition 2
       /dev/cciss/c1d1p3       Controller 1, disk 1, partition 3

Device c0d0p1 is controller 0 disk 0 partition 1 and if you created logical drives with the raid controller you should be seeing a device c0d1 and will need to create partitions on it.

On a separate note you said you were going to use RAID 0 (striping) on the last two drives and make it a NAS storage device. If people are going to be storing important information on that device you had better have good backups. RAID 0 is fast however the big drawback is that because data is split across the two drives, if either drive fails then you lose all of the data on both drives. Why not use all of the drives for one large RAID 5 array and then partition the drives for DATA and NAS. With RAID 5 if one of the drives fails the others know what is missing and the array can be rebuilt and you get the speed of striped data for both your data and NAS storage system. Just a suggestion..

Here is a link to a quick film about raid:
http://community.spiceworks.com/topic/457392-it-in-three-what-is-raid?swsrc=stumbleupon-community-signup-whiteboard-videos-what-is-raid-it-us

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.