Administrace komentářů

Jako poslední relikt, kde nemám btrfs jsou v domácím servříku 4 disky, které mají mdraid. A bohužel se mi v poslední době silně zpomalila kontrola na něm. Dříve spuštený raid-check (Centos distribuce, Debian distribuce má tuším checkarray s podobnou funkcionalitou) četl disky rychlostí cca 70-100MB/s v současnosti to dopadá takto.

cat /proc/mdstat
Personalities : [raid10] [raid6] [raid5] [raid4] 
md126 : active raid5 sdc2[1] sde2[4] sdd2[2] sdb2[0]
      7296182592 blocks super 1.2 level 5, 64k chunk, algorithm 2 [4/4] [UUUU]
      [=========>...........]  check = 48.5% (1180608128/2432060864) finish=1164.6min speed=17908K/sec
      bitmap: 0/19 pages [0KB], 65536KB chunk

md127 : active raid10 sde1[4] sdd1[2] sdb1[0] sdc1[1]
      995885056 blocks super 1.2 128K chunks 2 far-copies [4/4] [UUUU]
        resync=DELAYED
      bitmap: 0/8 pages [0KB], 65536KB chunk

jsem na nějakých 18MB/s, což je tedy dost děs. Přitom iostat je

iostat -xdmcz 1 
Linux 3.10.0-957.5.1.el7.x86_64 (gondor.doma)   02/10/2019      _x86_64_        (2 CPU)

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
          19.21   10.85   17.40    6.10    0.00   46.43

Device:         rrqm/s   wrqm/s     r/s     w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
sda               1.61     6.76   11.59   14.50     0.23     0.18    32.50     1.03   39.51    3.72   68.13   1.72   4.49
sdb            3066.47    14.55  139.10    8.45    13.60     0.21   191.69     1.07    7.23    2.98   77.13   1.53  22.54
sdc            3065.58    14.60  141.17    8.69    13.60     0.21   188.80     1.14    7.60    3.09   80.82   1.53  22.90
sdd            3067.27    14.61  141.12    8.62    13.60     0.21   188.91     1.10    7.34    3.11   76.48   1.53  22.95
sde            3067.21    14.68  138.77    8.35    13.58     0.21   192.02     3.61   24.51    4.90  350.38   1.94  28.51
dm-0              0.00     0.00   12.78   20.23     0.22     0.18    24.92     1.29   38.94    4.60   60.65   1.34   4.43
dm-1              0.00     0.00    0.42    1.04     0.00     0.00     8.05     0.01    8.21    7.75    8.40   0.57   0.08
dm-2              0.00     0.00    0.42    1.04     0.00     0.00     8.04     0.02   10.49    8.24   11.41   0.64   0.09
dm-3              0.00     0.00   12.78   19.97     0.22     0.18    25.12     4.96  151.40    5.24  244.94   1.56   5.12
md127             0.00     0.00  165.18   44.37     4.36     0.42    46.77     0.00    0.00    0.00    0.00   0.00   0.00
md126             0.00     0.00   22.92    0.56     0.64     0.00    56.42     0.00    0.00    0.00    0.00   0.00   0.00
dm-4              0.00     0.00   22.92    0.54     0.64     0.00    56.46     0.35   15.12    9.97  232.80   3.51   8.24
dm-5              0.00     0.00  165.18   44.31     4.36     0.42    46.79    39.64  189.13   11.72  850.55   1.25  26.23

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
          50.00    0.00   15.10    0.00    0.00   34.90

Device:         rrqm/s   wrqm/s     r/s     w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
sdb            9052.00     0.00  295.00    2.00    36.75     0.00   253.44     0.34    1.21    1.15   10.00   1.12  33.20
sdc            9053.00     0.00  293.00    2.00    36.57     0.00   253.89     0.35    1.18    1.13    8.50   1.15  33.90
sdd            9052.00     0.00  293.00    2.00    36.56     0.00   253.86     0.36    1.21    1.15    9.00   1.19  35.20
sde            9053.00     0.00  293.00    2.00    36.62     0.00   254.27     0.39    1.33    1.18   22.00   1.32  39.00
md126             0.00     0.00    7.00    0.00     0.38     0.00   109.71     0.00    0.00    0.00    0.00   0.00   0.00
dm-4              0.00     0.00    7.00    0.00     0.38     0.00   109.71     0.06    8.29    8.29    0.00   1.86   1.30

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
          17.35   29.08   11.73    0.00    0.00   41.84

Device:         rrqm/s   wrqm/s     r/s     w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
sda               0.00     0.00    0.00   13.00     0.00     0.16    25.23     0.00    0.23    0.00    0.23   0.23   0.30
sdb            2573.00     0.00   84.00    0.00    10.44     0.00   254.48     0.12    1.43    1.43    0.00   1.30  10.90
sdc            2574.00     0.00   84.00    0.00    10.50     0.00   255.90     0.12    1.40    1.40    0.00   1.32  11.10
sdd            2573.00     0.00   85.00    0.00    10.50     0.00   252.99     0.09    1.07    1.07    0.00   1.07   9.10
sde            2574.00     0.00   84.00    0.00    10.44     0.00   254.57     0.10    1.18    1.18    0.00   1.17   9.80
dm-0              0.00     0.00    0.00   13.00     0.00     0.16    25.23     0.00    0.23    0.00    0.23   0.23   0.30
dm-3              0.00     0.00    0.00   13.00     0.00     0.16    25.23     0.01    1.08    0.00    1.08   1.08   1.40
md126             0.00     0.00    7.00    0.00     0.38     0.00   109.71     0.00    0.00    0.00    0.00   0.00   0.00
dm-4              0.00     0.00    7.00    0.00     0.38     0.00   109.71     0.04    6.00    6.00    0.00   2.00   1.40

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           0.51   47.45    8.16    0.00    0.00   43.88

Device:         rrqm/s   wrqm/s     r/s     w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
sdb            2077.00     0.00   67.00    0.00     8.38     0.00   256.00     0.08    1.16    1.16    0.00   1.16   7.80
sdc            2077.00     0.00   67.00    0.00     8.38     0.00   256.00     0.08    1.24    1.24    0.00   1.24   8.30
sdd            2077.00     0.00   67.00    0.00     8.38     0.00   256.00     0.09    1.36    1.36    0.00   1.36   9.10
sde            2077.00     0.00   67.00    0.00     8.38     0.00   256.00     0.08    1.16    1.16    0.00   1.16   7.80

a nevidím zde nic, co by mělo situaci limitovat. Podobně atop dává

ATOP - gondor                                 2019/02/10  18:32:33                                 --------------                                 10s elapsed
PRC | sys    2.70s | user   0.23s | #proc    186  | #trun      2 | #tslpi   253 | #tslpu     1 | #zombie    0 | clones     2  |              | #exit      2 |
CPU | sys      23% | user      2% | irq       1%  | idle    173% | wait      1% |              | steal     0% | guest     0%  | curf  800MHz | curscal  50% |
cpu | sys      16% | user      1% | irq       0%  | idle     83% | cpu000 w  0% |              | steal     0% | guest     0%  | curf  800MHz | curscal  50% |
cpu | sys       7% | user      2% | irq       0%  | idle     91% | cpu001 w  1% |              | steal     0% | guest     0%  | curf  800MHz | curscal  50% |
CPL | avg1    1.59 | avg5    1.62 | avg15   1.60  |              |              | csw     9682 | intr    8886 |               |              | numcpu     2 |
MEM | tot     3.3G | free  152.9M | cache   2.2G  | buff  420.9M | slab  207.3M | shmem  25.0M | shrss   1.2M | vmbal   0.0M  | hptot   0.0M | hpuse   0.0M |
SWP | tot     4.0G | free    3.8G |               |              |              |              |              |               | vmcom   1.7G | vmlim   5.7G |
LVM |     uloziste | busy      1% | read      70  | write      0 | KiB/r     54 | KiB/w      0 | MBr/s    0.4 | MBw/s    0.0  | avq     4.69 | avio 1.80 ms |
LVM | 4bec-b9a1-1d | busy      1% | read       0  | write     85 | KiB/r      0 | KiB/w      5 | MBr/s    0.0 | MBw/s    0.0  | avq    10.50 | avio 1.19 ms |
LVM | cl_gondor-01 | busy      1% | read       0  | write     87 | KiB/r      0 | KiB/w      5 | MBr/s    0.0 | MBw/s    0.0  | avq     5.00 | avio 1.02 ms |
MDD |        md126 | busy      0% | read      70  | write      0 | KiB/r     54 | KiB/w      0 | MBr/s    0.4 | MBw/s    0.0  | avq     0.00 | avio 0.00 ms |
DSK |          sdc | busy     17% | read    1266  | write      2 | KiB/r    127 | KiB/w      2 | MBr/s   15.8 | MBw/s    0.0  | avq     1.03 | avio 1.29 ms |
DSK |          sdd | busy     17% | read    1271  | write      2 | KiB/r    127 | KiB/w      2 | MBr/s   15.8 | MBw/s    0.0  | avq     1.05 | avio 1.29 ms |
DSK |          sdb | busy     17% | read    1271  | write      2 | KiB/r    127 | KiB/w      2 | MBr/s   15.8 | MBw/s    0.0  | avq     1.07 | avio 1.28 ms |
DSK |          sde | busy     16% | read    1266  | write      2 | KiB/r    127 | KiB/w      2 | MBr/s   15.8 | MBw/s    0.0  | avq     1.00 | avio 1.23 ms |
DSK |          sda | busy      1% | read       0  | write     12 | KiB/r      0 | KiB/w     37 | MBr/s    0.0 | MBw/s    0.0  | avq     1.12 | avio 7.42 ms |
NFS | rpc      120 | cread    120 | cwrit      0  | MBcr/s   0.4 | MBcw/s   0.0 | nettcp   120 | netudp     0 | badfmt     0  | badaut     0 | badcln     0 |
NET | transport    | tcpi     630 | tcpo    2896  | udpi       7 | udpo       7 | tcpao      0 | tcppo      1 | tcprs      0  | tcpie      0 | udpie      0 |
NET | network      | ipi      637 | ipo     2903  | ipfrw      0 | deliv    637 |              |              |               | icmpi      0 | icmpo      0 |
NET | enp4s0    0% | pcki     631 | pcko    2897  | sp 1000 Mbps | si   54 Kbps | so 3314 Kbps | erri       0 | erro       0  | drpi       0 | drpo       0 |
NET | lo      ---- | pcki       6 | pcko       6  | sp    0 Mbps | si    2 Kbps | so    2 Kbps | erri       0 | erro       0  | drpi       0 | drpo       0 

nic není na limitu, nic není červené. Předpokládám, že se check zpomalil díky nějaké aktualizaci. Jediné na, co jsem přišel je, že v tree zobrazení htop je crond->bash raid-check->sleep 60. Takže to vypadá že skript volá sleep, ale netuším proč. (přihodil jsem skript jako přílohu) Hlavně nevím, co se změnilo, protože jsem porovnal skript s půl roku starou archivní zálohou a je zcela stejný, a v té době zcela jistě jela kontrola rychle.