I have new FAS2020 with Dual Controller fully populated with 12x 300GB SAS disks. I want know the best configuration option for me to make maximum available storage space for data storage.
by default the unit has two aggregates with two root volumes which is consuming 4 disks each.
i.e., With default Raid DP configuration one data disk, two parity disk and one spare disk is allocated for each aggregate. Hence rest available for data is only 4 disks. Which if do not create separate aggregate for data volumes will give me maximum of 1 TB usable space.
Can anyone suggest a solution to increase disk usability
Thanks in advance
Once i configured this configuration with the second head with a raid4 config for non production data like an iso store for vmware and all kind of software installtion packages or even a snapmirror. The primary was used for production. In this way the disk you save at the second head (raid4 uses 3 disk instead of raiddp 4, was added at the primary head. Now a failover is still posiible when a head failes.
You must have an root volume on the second head. I just rebooted the second head and pressed something at boot for the special boot menu (cant remeber now) and choose to format the head with a raid4 config. thats all tyou need to do. with this option you save 1 disk and you can add it to the first head for production data.
I've got a FAS2020 with pretty much the same specs. I use Controller1 for production and Controller2 for testing and ISOs like someone else in the thread mentioned.
Controller2 has a RAID4 volume comprised of only 2 disks. I do not have a hotspare configured for that aggregate. I have vol0 sized down to around 8GB on that controller. You can change a RAID-DP aggregate to RAID-4 without wiping out the aggregate, search to documentation or call your SE. Also when you only have 2 drives in the RAID4 it will give you a warning about not having a hotspare, but this is just for test/temp storage so who cares, it won't actually do anything.
Controller1 has the remaining 10 disks assigned to it in RAID-DP with a hot spare so I've got 9 disks in the aggregate. Again here I have vol0 sized down to around 10GB and use the remaining space in the aggr to create my flexvols.
This gives me a Active/Active config so that Controller2 can take over Controller1 if there is a problem or for NDUs.
To change the aggregate on one head to RAID4 from RAID-DP, use
aggr options aggr0 raidtype raid4
you will get the second parity disk as non-zeroed spare.
Now reassign this disk (and the other spare) to the second head:
disk remove_ownership ... (on the first head)
disk assign ... (on the second head)
Now sysconfig should only report 2 disks used. To disable the constant warnings about missing spare disks use the option raid.min_spare_count:
options raid.min_spare_count 0
you now have 10 disks total for the "production" head, which means 7 data + 2 Parity + Spare (i.e. about 2TB with 300g disks)
I don't believe you can set a RAID4 disk group to not warn about spares:
options raid.min_spare_count 0
option raid.min_spare_count: Cannot be set to 0 if there are more than
16 drives or any RAID4 groups configured.
For testing purpose you can create a flexible root volume using the special boot menu option, You will get the remaining aggregate disk space used for other volumes as well.