本文详细介绍了如何使用各种工具来监控和分析系统的性能，包括CPU使用率、内存利用率、锁竞争、抢占式上下文切换、网络I/O、磁盘I/O等关键指标，并通过实例展示了各工具的输出解读，帮助开发者深入了解系统运行状态。
一、CPU 使用率
命令： vmstat
输出
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
0 0 66972 2722752 453788 2639820 0 0 0 7 1 0 4 1 95 0 0 最后的CPU列:：us = user， sy=system ，id=idle ，wa=waiting ，st=
命令： mpstat
输出
Linux 2.6.32-431.23.3.el6.x86_64 (VA128) 12/28/2014 _x86_64_ (8 CPU)
01:18:49 PM CPU %usr %nice %sys %iowait %irq %soft %steal %guest %idle
01:18:49 PM all 4.30 0.00 0.70 0.03 0.00 0.02 0.00 0.00 94.94命令： top
输出
top - 13:20:33 up 27 days, 2:14, 4 users, load average: 0.01, 0.01, 0.12
Tasks: 367 total, 1 running, 366 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.9%us, 1.8%sy, 0.0%ni, 97.2%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 16333648k total, 15140648k used, 1193000k free, 453800k buffers
Swap: 6291448k total, 66972k used, 6224476k free, 2640100k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
17253 admin 20 0 1340m 257m 20m S 2.3 1.6 306:08.27 phMonitor
611 root 20 0 0 0 0 S 2.0 0.0 20:55.19 vmmemctl
4683 admin 20 0 2842m 1.9g 9780 S 2.0 12.4 6299:58 java
27 root RT 0 0 0 0 S 1.7 0.0 0:57.62 migration/6
6340 admin 20 0 6255m 83m 11m S 1.0 0.5 0:29.15 jstatd
6461 admin 20 0 8262m 1.4g 27m S 0.7 8.7 34:09.20 java
9431 admin 20 0 4432m 3.7g 532m S 0.7 23.6 152:41.77 phQueryWorker
14706 admin 20 0 15268 1548 1004 R 0.7 0.0 0:00.14 top
730 admin 20 0 1271m 750m 532m S 0.3 4.7 215:21.15 phRuleWorker
744 admin 20 0 915m 559m 532m S 0.3 3.5 106:13.52 phIpIdentityWor
3082 root 20 0 134m 8108 1124 S 0.3 0.0 44:56.82 redis-server 二、CPU调度程序运行队列
命令： vmstat
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
2 0 36752 788396 431420 2173544 0 0 0 7 0 1 20 3 77 0 0 三、内存利用率
命令： vmstat
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
2 0 36752 788396 431420 2173544 0 0 0 7 0 1 20 3 77 0 0 memory 项
命令：free
输出：
total used free shared buffers cached
Mem: 16333656 15563320 770336 0 431452 2172656
-/+ buffers/cache: 12959212 3374444
Swap: 6291448 36752 6254696
命令： top
输出
top - 13:20:33 up 27 days, 2:14, 4 users, load average: 0.01, 0.01, 0.12
Tasks: 367 total, 1 running, 366 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.9%us, 1.8%sy, 0.0%ni, 97.2%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 16333648k total, 15140648k used, 1193000k free, 453800k buffers
Swap: 6291448k total, 66972k used, 6224476k free, 2640100k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
17253 admin 20 0 1340m 257m 20m S 2.3 1.6 306:08.27 phMonitor
611 root 20 0 0 0 0 S 2.0 0.0 20:55.19 vmmemctl
4683 admin 20 0 2842m 1.9g 9780 S 2.0 12.4 6299:58 java
27 root RT 0 0 0 0 S 1.7 0.0 0:57.62 migration/6
6340 admin 20 0 6255m 83m 11m S 1.0 0.5 0:29.15 jstatd
6461 admin 20 0 8262m 1.4g 27m S 0.7 8.7 34:09.20 java
9431 admin 20 0 4432m 3.7g 532m S 0.7 23.6 152:41.77 phQueryWorker
14706 admin 20 0 15268 1548 1004 R 0.7 0.0 0:00.14 top
730 admin 20 0 1271m 750m 532m S 0.3 4.7 215:21.15 phRuleWorker
744 admin 20 0 915m 559m 532m S 0.3 3.5 106:13.52 phIpIdentityWor
3082 root 20 0 134m 8108 1124 S 0.3 0.0 44:56.82 redis-server 四、锁竞争、抢占式上下文切换
命令：pidstat -w
输出：
Linux 2.6.32-431.el6.x86_64 (wk129) 12/28/2014 _x86_64_ (8 CPU)
01:34:57 PM PID cswch/s nvcswch/s Command
01:34:57 PM 1 0.00 0.00 init
01:34:57 PM 2 0.00 0.00 kthreadd
01:34:57 PM 3 0.49 0.00 migration/0
01:34:57 PM 4 1.19 0.00 ksoftirqd/0
01:34:57 PM 5 0.00 0.00 migration/0
01:34:57 PM 6 0.08 0.00 watchdog/0
01:34:57 PM 7 0.45 0.00 migration/1
01:34:57 PM 8 0.00 0.00 migration/1
01:34:57 PM 9 0.50 0.00 ksoftirqd/1
01:34:57 PM 10 0.08 0.00 watchdog/1
01:34:57 PM 11 0.42 0.00 migration/2
01:34:57 PM 12 0.00 0.00 migration/2
01:34:57 PM 13 0.50 0.00 ksoftirqd/2
01:34:57 PM 14 0.08 0.00 watchdog/2
01:34:57 PM 15 0.40 0.00 migration/3
01:34:57 PM 16 0.00 0.00 migration/3
01:34:57 PM 17 0.45 0.00 ksoftirqd/3
01:34:57 PM 18 0.08 0.00 watchdog/3
01:34:57 PM 19 0.41 0.00 migration/4
01:34:57 PM 20 0.00 0.00 migration/4
01:34:57 PM 21 0.47 0.00 ksoftirqd/4
01:34:57 PM 22 0.08 0.00 watchdog/4
01:34:57 PM 23 0.40 0.00 migration/5
01:34:57 PM 24 0.00 0.00 migration/5
01:34:57 PM 25 0.53 0.00 ksoftirqd/5
01:34:57 PM 26 0.08 0.00 watchdog/5
01:34:57 PM 27 0.44 0.00 migration/6
01:34:57 PM 28 0.00 0.00 migration/6
01:34:57 PM 29 0.53 0.00 ksoftirqd/6
01:34:57 PM 30 0.08 0.00 watchdog/6
01:34:57 PM 31 0.40 0.00 migration/7
01:34:57 PM 32 0.00 0.00 migration/7
01:34:57 PM 33 0.46 0.00 ksoftirqd/7
01:34:57 PM 34 0.08 0.00 watchdog/7
01:34:57 PM 35 1.91 0.00 events/0
01:34:57 PM 36 1.08 0.00 events/1
01:34:57 PM 37 1.27 0.00 events/2五、网络I/O
命令：nicstat
六、磁盘I/O
命令： iostat -xc
输出：
Linux 2.6.32-431.el6.x86_64 (wk129) 12/28/2014 _x86_64_ (8 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
19.83 0.00 3.21 0.17 0.00 76.78
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util
sda 0.01 6.24 0.07 4.63 2.31 87.14 19.04 0.75 159.79 3.18 1.49
sdb 0.00 2.34 0.00 0.37 0.00 21.73 58.31 0.01 25.83 4.00 0.15
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
