nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Getting the most out of my disks

joesmith

Hello,

I am somewhat new to the storage game and working for a new company that recently bought a FAS2750 and 1 additional disk shelf.

I am trying to get our system set up in the best way and can't seem to get use out a large chunk of our disks.

The FAS2750 has 24 1.64TB disks.

the Disk shelf has 12 7.15TB disks.

When I look at my disk summary, it says that i have 85.67TB of spare disk space.

As far i understand I would need to place the disk shelf disks into a different aggregate because the are different sized disks from the ones in the controller unit. I can't seem to figure out how to get the existing aggregates to let go of these disks so that I can add them to their own aggregate. Is that even the best thing to do with the spare disks?

Please let me know if I should elaborate on anything.

Thanks,

Joe

Find more posts tagged with

aggregates

Accepted answers

SpindleNinja

1.1.0 thought 1.1.11 are spares, so no need to delete any aggrs.

Have you verified that the new shelf cabling is good using something like Config Advisor?

Nodes (controllers) own disks, and disks get pooled to become aggrs.

for this reconfigure you'd want to do something like this:

note: there are other ways to do this using wildcards * and the -force flag. but i figured this is the safest way for a first time doing this.

storage disk options modify -autoassign off

storage disk removeowner -disk 1.1.0

storage disk removeowner -disk 1.1.2

storage disk removeowner -disk 1.1.4

storage disk removeowner -disk 1.1.6

storage disk removeowner -disk 1.1.8

storage disk removeowner -disk 1.1.10
storage disk assign -disk 1.1.0 -owner node-01

storage disk assign -disk 1.1.2 -owner node-01

storage disk assign -disk 1.1.4 -owner node-01

storage disk assign -disk 1.1.6 -owner node-01

storage disk assign -disk 1.1.8 -owner node-01

storage disk assign -disk 1.1.10 -owner node-01
storage disk show

storage disk options modify -autoassign on

at this point all the new 8TB SATA drives will be owned by node-01 and you can create a new AGGR 11 drives + spare.

All comments

SpindleNinja

With that config, I typically do all the "internal" (1.64T drives) aggr for controller 01. And the 1.8 drives as an aggr (+1 spare) on controller 02. It's not a 50/50 match, but it's probably the simplest way to do it.

There's other options though. You can mix drives inside an aggr. So you can have 1 raid group of 1.67 drives and one of 1.8. but that would make it all on one controller.

paul_stejskal

He said 7 TB drives.

You'll have to use two of those at a minimum with RAID DP for parity, or three with RAID TEC.

SpindleNinja

i swear that said 1.8 sorry about that OP.

joesmith

Thank you for you input guys.

I understand that I am going to have lose the space on two or three of the disks for DP but how do i remove these 7TB disks from the aggrs that they are in right now to create a new aggregate?

Or am I able to just keep them in the aggregate they are in but use them for Data instead of spares? <--- This would be "No" right? because you can't mix disk size in the aggregates?

Thanks,

Joe