Hello.
I have setup an array of 7x3tb WD30EZRX drives, though its meant for 8x,
so it runs in singledegraded mode.
the issue is that i get very poor performance, generally roughly 25MB/s
writes only. individually the disks are fine.
iowait and idle is high:
avg-cpu: %user %nice %system %iowait %steal %idle
2.29 0.00 5.85 22.14 0.00 69.72
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s
avgrq-sz avgqu-sz await svctm %util
sdb 1370.00 732.00 109.00 30.00 11824.00 5864.00
127.25 1.50 10.71 2.45 34.00
sdf 1392.00 839.00 88.00 33.00 11840.00 6976.00
155.50 3.42 28.26 4.43 53.60
sdd 1422.00 863.00 64.00 29.00 13448.00 6384.00
213.25 2.28 34.54 4.56 42.40
sdg 1388.00 446.00 92.00 18.00 11840.00 3248.00
137.16 0.63 5.71 1.60 17.60
sdc 1395.00 857.00 85.00 40.00 11840.00 6944.00
150.27 1.07 8.58 1.86 23.20
sda 1370.00 985.00 111.00 41.00 11888.00 7744.00
129.16 5.30 35.79 5.21 79.20
sde 1417.00 669.00 70.00 21.00 13528.00 5040.00
204.04 1.94 32.79 4.53 41.20
sdh 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00
md0 0.00 0.00 0.00 86.00 0.00 17656.00
205.30 0.00 0.00 0.00 0.00
dm-0 0.00 0.00 0.00 0.00 0.00 0.00
0.00 4174.84 0.00 0.00 100.00
http://paste.kde.org/547370/ - the same as above, but on pastebin since
might be annoying to read in mail client depending on font used.
(sdh is not part of array)
mdadm detail:
/dev/md0:
Version : 1.2
Creation Time : Sat Sep 8 23:01:11 2012
Raid Level : raid6
Array Size : 17581590528 (16767.11 GiB 18003.55 GB)
Used Dev Size : 2930265088 (2794.52 GiB 3000.59 GB)
Raid Devices : 8
Total Devices : 7
Persistence : Superblock is persistent
Update Time : Wed Sep 12 01:55:49 2012
State : active, degraded
Active Devices : 7
Working Devices : 7
Failed Devices : 0
Spare Devices : 0
Layout : left-symmetric
Chunk Size : 512K
Name : mainserver:0 (local to host mainserver)
UUID : d48566eb:ca2fce69:907602f4:84120ee4
Events : 26413
Number Major Minor RaidDevice State
0 8 0 0 active sync /dev/sda
1 8 16 1 active sync /dev/sdb
2 8 32 2 active sync /dev/sdc
3 8 48 3 active sync /dev/sdd
4 8 64 4 active sync /dev/sde
5 8 80 5 active sync /dev/sdf
8 8 96 6 active sync /dev/sdg
7 0 0 7 removed
It should be noted that I tested the chunksizes extensively, from 4k to
2048k, and the default seemed to offer best performance allround, very
close to the best performers for all workloads, and much much better
than the worst.
I conducted tests with dd directly on md0, and xfs on md0, and with
dm-crypt on top md md0, both dd and xfs on the dm-0 device. resulting in
marginal performance difference, so i must assume the issue is in the
raid layer.
kernel is:
Linux mainserver 3.2.0-0.bpo.1-amd64 #1 SMP Sat Feb 11 08:41:32 UTC 2012
x86_64 GNU/Linux
could this be due to being in singledegraded mode? after im done copying
files to this array i will be adding the last disk.
Any input is welcomed.
Oh, and im not subscribed to the list, so please CC me.
--
Kasper Sandberg
Sandberg Enterprises
+45 51944242
http://www.sandbergenterprises.dk
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
