I just installed Rocky 9 on a newly acquired HP Proliant Dl360p. My first attempt at a server. I used 4 drives in an lvm and left 4 drives as spare with the intention of using them for storage independent of the lvm used for the system. I have made many searches but can’t find any information addressing my situation. Any and all help will be greatly appreciated. Thanks.
The first question is what block devices does your system see? You get answer to that with lsblk
The default output format of lsblk does not tell everything. IMHO, more useful columns come with:
lsblk -o name,size,type,fstype,uuid,mountpoints
A disk has partition(s), partition has filesystem, and OS mounts filesystem to mountpoint.
The rules of what to mount where are mainly in /etc/fstab.
Look at that file and the output of lsblk; The {uuid, mountpoint, fstype} triplets should match.
In your case some partition(s) do not have filesystem directly. Partition has PV – LVM physical volume.
One or more PV belong to a VG – LVM volume group.
Within VG there are LVs – LVM logical volumes. A LV contains a filesystem.
While the lsblk tells quite a lot, more LVM details you can get with pvs, vgs, and lvs.
You said that you did use 4 drives. You probably do now have at least one filesystem that spans multiple drives (because LVM makes that possible). There comes the risk that what if one of the drives breaks? (Being in one drive has that same risk; the drive can break.) A usual mitigation is to have reduncancy, e.g. RAID-1 (aka mirror), RAID-5, or RAID-6. LVM might offer some equivalents to those. Redundancy obviously reduces space that is available for data.
That said, a basic procedure would be to create partitioning (with ‘gdisk’), create filesystem(s) (with ‘mkfs’), and update /etc/fstab to mount the new filesystems on boot. If you want RAID and/or LVM, then there are additional steps (before creation of filesystem).
I do not agree jlehtone.
The first question should be - Will a crystal ball be enough to know the full specification of HP Proliant Dl360p and generation.
And then if there is a RAID controller (probably there is) how is it configured to present to OS the disk.
And only then can one guess where they are lost and/or why.
You are on the right path. I didn’t want to use the hardware controller but didn’t know how to get around that. I have 8 drives. I used the hardware controller to make a logical drive from the first 4 drives hoping that the other 4 would be available for using them separately. I haven’t loaded anything of consequence so I can start over with some guidance.
HP ProLiant DL360p Gen8 1U RackMount 64-bit Server with 2×6-Core E5-2640 Xeon 2.5GHz CPUs + 64GB PC3-10600R RAM + 8×300GB 10K SAS SFF HDD, P420i RAID, 4×GigaBit NIC, 2×Power Supplies
To use the other disks individually, you will need to configure that under the hardware controller accordingly as well. I have a DL360 Gen8 and also had to do something similar recently. So you basically just select one disk and repeat 4 times instead of selecting multiple disks to create a raid config.
Want to bypass the controller?
And why is money given for it?
Once available use it.
You say you are learning and this is your first server.
Okay, but at least theoretically learn what it is “redundant array of independent disks” (RAID)
Then your available controller
That way you’ll know what you’re doing. I.e. you know what you want to happen, and you will try to make it happen.
Personally, I don’t like HP servers as a closed configuration.
How can I help from the information provided?
SAS drives without a controller are basically impossible to use in this MB and chassis.
By default, additional spare disks associated with the array are invisible and inaccessible from OS
with 8 disks you can make the following configurations:
RAID1, RAID1+E, RAID5, probably RAID6 or RAID10
After BIOS select and enter the controller interface.
Configure the disks and associate them in the array with your preferred configuration.
(for example 6 disks = RAID6 + 2 disk hot spare)