ProLiant Servers (ML,DL,SL)
Showing results for 
Search instead for 
Did you mean: 

RIAD and Linux Partision Size

Frequent Advisor

RIAD and Linux Partision Size

In a case that DL380 has 6 disks: Disk 1 and Disk 2 (RAID 1 implemented), Disks 3 to 5 (RAID 5 implemented), and the last disk as an online spare. The Operating System is Linux. The hardware RAID (HP smart array)is applied.

1) Suppose the Disks 1 and 2 are 72 GB EACH. Since both them are under RAID 1, when I do the disk partision, what are the size (storage capacity) available, 72 GB or 2x72 GB?

2) Suppose the Disks 3, 4 and 5 are 146 GB EACH. Since all three of them are under RAID 5, when I do the disk partision, what are the size (storage capacity) available, 2x146 GB or 3x146 GB?

3)What about the Disk #6 (online spare), should I just leave it there without the partision? On the event that a disk in either RAID 1 or RAID 5 crashed, this spare disk will be automatically brought into opration and the necessary partition will be done "automatically"--Is this description correct? (Otherwise, I really do not know how the Disk #6 should be partisioned beforehand, since each array has its own partition requirements)

Thanks to help.


Igor Karasik
Honored Contributor

Re: RIAD and Linux Partision Size

1) size will be 72 Gb
2) size will be 2*146 Gb
3) You add disk 6 as spare drive for both arrays and you don't need partition
on disk 6.
You can see this "old" HP slideshow for RAID understanding:
Honored Contributor

Re: RIAD and Linux Partision Size


Just one thing to add to Igor's reply.

Disk 6 must be 146Gb since it must act as a spare for the raid 5 and the raid 1 volumes.
You can asssign a spare drive to more than one array but it has to be the same or larger size than the largest disk members in the array.

If it was to be used as a replacement for a failed drive in the raid1 set (72Gb members), it will adopt the geometry of a 72Gb drive. It is therefore important that if this disk gets used at some point for the raid 1, you will need to get a 72Gb replacement for the faulty disk and put it back into the original raid 1 set to free up the 146Gb spare again.

I hope this makes sense.