Prepping to install: a different tangent

["William A. Mahaffey III" <wam%hiwaay.net@localhost>]

In the course of my breath-takingly documented efforts to get my little 
server installed w/ NetBSD, I had reason/need to redo the RAID5 setup 
twice. I attach the 2 'raid.conf' files. Both times, the raid devices 
were created from 5 slices which were/are each identically sized at 
about 950 GiB. The 1st time, I used the 1st raid.conf & didn't specify a 
spare drive, I figured/fantasized that raidctl would handle that itself. 
That raid device formatted out to about 3.6 T-B, & took a touch over 5 
hours to write out parity data. That was about the expected size of a 
5-element RAID5 device, w/ 4 active devices & 1 spare, so I was happy. 
More reading of the online man pages convinced me that I did indeed need 
to specify a spare device, so the 2nd time I used the 2nd raid.conf to 
setup from the same 5 slices. That time, it took almost 7 hours to write 
out the parity data & formatted out to about 2.7 TiB, which sounds to me 
like 3 active devices & 2 spares. So, the question is: Do I indeed have 
to specify a spare drive explicitly in my raid.conf, or will raidctl 
intelligently decide to pick 1 & setup 'N-1' active devices for 'N' 
specified in the raid.conf file ? I also attach the output for 'raidctl 
-s' after the 2nd setup, which looks right, but raises another question: 
does a RAID5 setup really eat up as much storage as it looks like, i.e. 
4 active ~950 GiB slices only yield about 2.7 TiB of usable space ? TIA 
& thanks for the patience & info so far :-) ....


--

	William A. Mahaffey III

 ----------------------------------------------------------------------

	"The M1 Garand is without doubt the finest implement of war
	 ever devised by man."
                           -- Gen. George S. Patton Jr.

# RAID5 on 5 (4+1) partitions, 32K blocksize ....

# row col spare
START array
1 5 0

START disks
/dev/wd0f
/dev/wd1f
/dev/wd2f
/dev/wd3f
/dev/wd4f

# secPerSU SUsPerParityUbit SUsPerReconUnit RAID-level
START layout
16 1 1 5

START queue
fifo 100

# RAID5 on 5 (4+1) partitions, 32K blocksize ....

# row col spare
START array
1 4 1

START disks
/dev/wd0f
/dev/wd1f
/dev/wd2f
/dev/wd3f

START spare
/dev/wd4f

# secPerSU SUsPerParityUbit SUsPerReconUnit RAID-level
START layout
16 1 1 5

START queue
fifo 100

Components:
           /dev/wd0f: optimal
           /dev/wd1f: optimal
           /dev/wd2f: optimal
           /dev/wd3f: optimal
Spares:
           /dev/wd4f: spare
Component label for /dev/wd0f:
   Row: 0, Column: 0, Num Rows: 1, Num Columns: 4
   Version: 2, Serial Number: 1234567, Mod Counter: 31
   Clean: No, Status: 0
   sectPerSU: 16, SUsPerPU: 1, SUsPerRU: 1
   Queue size: 100, blocksize: 512, numBlocks: 1886414192
   RAID Level: 5
   Autoconfig: Yes
   Root partition: No
   Last configured as: raid2
Component label for /dev/wd1f:
   Row: 0, Column: 1, Num Rows: 1, Num Columns: 4
   Version: 2, Serial Number: 1234567, Mod Counter: 31
   Clean: No, Status: 0
   sectPerSU: 16, SUsPerPU: 1, SUsPerRU: 1
   Queue size: 100, blocksize: 512, numBlocks: 1886414192
   RAID Level: 5
   Autoconfig: Yes
   Root partition: No
   Last configured as: raid2
Component label for /dev/wd2f:
   Row: 0, Column: 2, Num Rows: 1, Num Columns: 4
   Version: 2, Serial Number: 1234567, Mod Counter: 31
   Clean: No, Status: 0
   sectPerSU: 16, SUsPerPU: 1, SUsPerRU: 1
   Queue size: 100, blocksize: 512, numBlocks: 1886414192
   RAID Level: 5
   Autoconfig: Yes
   Root partition: No
   Last configured as: raid2
Component label for /dev/wd3f:
   Row: 0, Column: 3, Num Rows: 1, Num Columns: 4
   Version: 2, Serial Number: 1234567, Mod Counter: 31
   Clean: No, Status: 0
   sectPerSU: 16, SUsPerPU: 1, SUsPerRU: 1
   Queue size: 100, blocksize: 512, numBlocks: 1886414192
   RAID Level: 5
   Autoconfig: Yes
   Root partition: No
   Last configured as: raid2
/dev/wd4f status is: spare.  Skipping label.
Parity status: clean
Reconstruction is 100% complete.
Parity Re-write is 100% complete.
Copyback is 100% complete.