26.06.2011 19:26, Vadim Goncharov пишет:
26.06.11 @ 20:09 Vladislav V. Prodan wrote:

Как определить узкое место?
У машины после перезагрузки выросли interrupt до 40-60%

top -SH
last pid: 71672; load averages: 0.08, 0.04, 0.01 up 1+14:17:00 02:09:04
290 processes: 3 running, 271 sleeping, 16 waiting
CPU:  0.0% user,  0.0% nice,  0.0% system, 25.0% interrupt, 75.0% idle
Mem: 2564M Active, 225M Inact, 788M Wired, 66M Cache, 272M Free
Swap: 32G Total, 536M Used, 31G Free, 1% Inuse

  PID USERNAME      PRI NICE   SIZE    RES STATE   C   TIME   WCPU COMMAND
11 root 171 ki31 0K 32K CPU0 0 35.1H 95.17% {idle: cpu0} 12 root -32 - 0K 256K WAIT 1 11:13 75.88% {swi4: clock} 11 root 171 ki31 0K 32K RUN 1 34.4H 14.79% {idle: cpu1}
91863 mysql          57    0  3267M  2447M ucond   1   0:10  4.79% {mysqld}
0 root -68 0 0K 1088K - 0 5:55 0.59% {dummynet}
71350 www            46    0   222M 18656K accept  0   0:02  0.59% php-fpm
91863 mysql          45    0  3267M  2447M ucond   0   0:10  0.29% {mysqld}
0 root -16 0 0K 1088K - 1 32:24 0.00% {zio_read_intr} 0 root -16 0 0K 1088K - 1 12:10 0.00% {zio_write_issue} 5 root -8 - 0K 60K tx->tx 0 10:31 0.00% {txg_thread_enter} 12 root -64 - 0K 256K WAIT 1 9:23 0.00% {irq22: atapci0}
    3 root           -8    -     0K    16K -       0   8:37  0.00% g_up
    4 root           -8    -     0K    16K -       0   4:50  0.00% g_down
12 root -28 - 0K 256K WAIT 0 4:43 0.00% {swi5: +}
 4022 root           44    0 20884K  2304K nanslp  1   1:33  0.00% {monit}
12 root -44 - 0K 256K WAIT 1 1:31 0.00% {swi1: netisr 0}
97220 root           44    0 10580K  1896K ttyin   1   1:17  0.00% systat
   21 root           20    -     0K    16K syncer  1   1:06  0.00% syncer
66218 www            44    0 47572K  6856K kqread  1   1:02  0.00% nginx
66219 www            44    0 47572K  7020K kqread  1   1:00  0.00% nginx
0 root -16 0 0K 1088K sched 1 0:43 0.00% {swapper} 5 root -8 - 0K 60K arc_re 1 0:35 0.00% {arc_reclaim_thre}
 2029 root           44    0 30632K  8452K select  1   0:30  0.00% bsnmpd


vmstat -i

# vmstat -i
interrupt                          total       rate
irq16: ohci0 ohci1                     2          0
irq18: ohci2 ohci3+                    4          0
irq22: atapci0                  36877247        267
cpu0: timer                    275656570       1999
irq256: re0                     33104196        240
cpu1: timer                    275656154       1999
Total                          621294173       4507


--
Vladislav V. Prodan
VVP24-UANIC
+380[67]4584408
+380[99]4060508
vla...@jabber.ru

Ответить