0

I'm have RAID 5 on Synology DS1813+ (DSM Version: DSM 5.0-4493 Update 4) with 8 ST4000DM000-1F2168 disk, usind as NAS and iSCSI/NFS targer for ESXi.

In low read/write rate from disk have more io wait, please have you any ideas?

SRV1> uname -a
Linux SRV1 3.2.40 #4493 SMP Thu Aug 21 21:46:06 CST 2014 x86_64 GNU/Linux synology_cedarview_1813+

nmon output

DiskName Busy  Read WriteKB
sda       63%    0.0    0.0
sda1       0%    0.0    0.0
sda2      28%  115.7    0.0
sda3      50%    0.0    0.0
sdb       50%    0.0    0.0
sdb1       0%    0.0    0.0
sdb2      20%   85.8    0.0
sdb3      37%    0.0    0.0
<skip omitted>
Totals Read-MB/s=0.7      Writes-MB/s=0.0      Transfers/sec=884.8

CPU output

CPU  User%  Sys% Wait% Idle
  1   0.5   2.5  76.9  20.1
  2   0.0   1.0  59.1  39.9
  3   0.0  19.1  18.1  62.8
  4   0.0   2.5  18.5  79.0

RAID info:

SRV1> mdadm --detail /dev/md2
/dev/md2:
        Version : 1.2
  Creation Time : Mon Jan 20 12:26:15 2014
     Raid Level : raid5
     Array Size : 27316073792 (26050.64 GiB 27971.66 GB)
  Used Dev Size : 3902296256 (3721.52 GiB 3995.95 GB)
   Raid Devices : 8
  Total Devices : 8
    Persistence : Superblock is persistent

    Update Time : Mon Jan 18 08:51:10 2016
          State : clean
 Active Devices : 8
Working Devices : 8
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 64K

           Name : SRV1:2  (local to host SRV1)
           UUID : 690db49f:cc6aebea:e4dc7a24:b8c4a142
         Events : 66

    Number   Major   Minor   RaidDevice State
       0       8        3        0      active sync   /dev/hda3
       1       8       19        1      active sync   /dev/hdb3
       2       8       35        2      active sync   /dev/hdc3
       3       8       51        3      active sync   /dev/sdd3
       4       8       67        4      active sync   /dev/hde3
       5       8       83        5      active sync   /dev/sdf3
       6       8       99        6      active sync   /dev/sdg3
       7       8      115        7      active sync   /dev/sdh3

In dmesg:

[23334402.188177] __fd_async_epio_request_handle: 6 callbacks suppressed
[23334402.194671] iSCSI: iSCSI: EPIO[/volume1/@iSCSITrg/iSCSI_3_Extent_LUN-L1_000], opcode[89], error(-65539), pos(10578606080)
[23334402.194937] iSCSI: iSCSI: EPIO[/volume1/@iSCSITrg/iSCSI_3_Extent_LUN-L1_000], opcode[89], error(-65539), pos(10578606080)
t3mp
  • 111
  • 2
  • I am not familiar with the Synology DS1813. Is `SRV1> mdadm --detail /dev/md2` from the unit's command line interface? It seems to be using software raid. – Gmck Jan 19 '16 at 14:24
  • Yep, it's software raid with 2x SiI 3132 controller. – t3mp Jan 19 '16 at 14:35
  • I am not sure if you can run these, or if it will affect the support/warranty, but what the the output of: `blockdev --getra /dev/md0` and `cat /sys/block/md0/md/stripe_cache_size` – Gmck Jan 19 '16 at 14:43
  • @Gmck: > blockdev --getra /dev/md0 8192 > cat /sys/block/md0/md/stripe_cache_size cat: can't open '/sys/block/md0/md/stripe_cache_size': No such file or directory – t3mp Jan 19 '16 at 15:16
  • You could try increasing the cache and see if that helps. If you can do this in the unit's admin interface, that is probably better. But if you want to try from the command line `blockdev --setra 20480 /dev/md0` If that does not help, contact the vendor and get them to troubleshoot further. – Gmck Jan 19 '16 at 15:37
  • @Gmck Thx for reply, this no help( – t3mp Jan 20 '16 at 07:25
  • Let us [continue this discussion in chat](http://chat.stackexchange.com/rooms/34599/discussion-between-t3mp-and-gmck). – t3mp Jan 20 '16 at 12:59

0 Answers0