Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

I've got an mdadm striped device (/dev/md0) from /dev/sdm{1..4}.

Here is an output of iostat 30 and then mdadm --detail on the mdadm device

# iostat 30
Linux 2.6.32-309-ec2 (ip-xx-xx-xx-xx)  01/28/2014      _x86_64_        (8 CPU)

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           8.35    0.24    3.39    3.42    0.02   84.58

Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read   Blk_wrtn
sda1              8.86        11.25       100.72     516762    4625936
sdb               0.00         0.02         0.00       1002        144
sdm3            254.04       554.86      8791.79   25483434  403790096
sdm4            258.81       554.71      8764.06   25476890  402516462
sdo               9.27        10.38       684.68     476565   31445952
sdm2            267.37       554.81     10149.08   25481144  466127624
sdm1            251.19       554.88      9336.52   25484630  428808386
md0            2157.06      2219.21     37041.77  101924170 1701257104
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           5.95    0.00    3.54    2.16    0.04   88.31

Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read   Blk_wrtn
sda1              1.70         0.00        13.60          0        408
sdb               0.00         0.00         0.00          0          0
sdm3            469.41       164.21     18973.14       4928     569384
sdm4            480.24       153.28     18557.81       4600     556920
sdo               2.90         0.00        64.25          0       1928
sdm2            509.00       138.35     22585.90       4152     677803
sdm1            467.64       157.28     20151.08       4720     604734
md0            5020.86       613.13     80597.97      18400    2418745


# mdadm --detail /dev/md0
/dev/md0:
        Version : 01.01
  Creation Time : Fri Mar  4 02:24:23 2011
     Raid Level : raid0
     Array Size : 419430144 (400.00 GiB 429.50 GB)
   Raid Devices : 4
  Total Devices : 4
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Fri Mar  4 02:24:23 2011
          State : clean
 Active Devices : 4
Working Devices : 4
 Failed Devices : 0
  Spare Devices : 0

     Chunk Size : 64K

           Name : xxxxxxx:0
           UUID : 48f20c12:6b08bf97:9b44ec7a:c0af2c68
         Events : 0

    Number   Major   Minor   RaidDevice State
       0       8      193        0      active sync   /dev/sdm1
       1       8      194        1      active sync   /dev/sdm2
       2       8      195        2      active sync   /dev/sdm3
       3       8      196        3      active sync   /dev/sdm4

fyi, each of the /dev/sd{1..4} volumes is a provisioned IOPS (500 iops) EBS device. Here is output of mdadm --detail.

I'm guessing it might be because they are using different block sizes or some kind of write back cache. Anyone explain this to me?

Thanks

share|improve this question

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.