写在前面

推送的的邮件里看到有大佬讲的公共课，听了之后这里整理学习笔记。
因为是公开课，所以讲的很浅，没接触过，这里做为了解，长长见识。

博文内容包括
系统调优原理概述
如何检测系统的性能瓶颈
如何进行内核参数调优
如何限制服务的资源占用
自定义tuned调优配置集

我突然又明白，死亡是聪明的兄长，我们可以放心地把自己托付给他，他会知道在我们有所准备的适当时刻前来。我也突然懂得，原来痛苦、失望和悲愁不是为了惹恼我们，使我们气馁或者无地自容；它们的存在，是为了使我们心智成熟，臻于完善。—赫尔曼·黑塞《彼得·卡门青》

零、系统调优原理概述：

通过调整系统设置来提高计算机资源的利用率、数据吞吐量或用戶体验的过程，消除或减轻系统中的排队现象。

分析系统性能的方法：

分析利用率、饱和、错误 (Utilization, Saturation and Errors (USE))和系统资源	–
资源 Resource	服务器中的硬件资源(CPU、内存、网络、磁盘和控制器)
利用率 Utilization	资源提供服务的平均时间（70%）
饱和度 Saturation	资源承担额外工作量的能力，这类工作通常排队(队列长度)
错误 Errors	错误事件数量

执行系统变更的步骤：

执行系统变更的步骤
设定基线：在调优中，执行性能测试，收集指标
执行变更：在系统上执行变更，建议查阅相应官方文档
验证变更有效性：执行变更后，重新运行测试，比较一些关键指标
取消变更并与基线对比：如有需要，可以取消变更并返回到基线状态
应用最终更改：应用并记录最终变更

一、如何检测系统的性能瓶颈

1、检测系统当前运行负载：vmstat、mpstat、iostat、top 等

装包，一般不需要

yum install -y sysstat

sysstat包含监测系统性能及效率的一组工具，这些工具对于我们收集系统性能数据，比如 CPU 使用率、硬盘和网络吞吐数据，这些数据的收集和分析，有利于我们判断系统是否正常运行，是提高系统运行效率、安全运行服务器的得力助手。

vmstat 内存子系统监控

┌──[root@liruilongs.github.io]-[~]
└─$ vmstat
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st1  0      0 3228884   2076 594152    0    0    10     6  102  152  0  0 100  0  0
┌──[root@liruilongs.github.io]-[~]
└─$ vmstat 1
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st1  0      0 3228852   2076 594184    0    0    10     6  102  152  0  0 100  0  0  # 平均值0  0      0 3228852   2076 594184    0    0     0     0  197  289  0  1 99  0  00  0      0 3228852   2076 594184    0    0     0     0  192  290  0  0 100  0  01  0      0 3228852   2076 594184    0    0     0     0  180  285  0  0 100  0  00  0      0 3228852   2076 594184    0    0     0     0  174  276  1  0 99  0  00  0      0 3228852   2076 594184    0    0     0     0  173  267  0  0 100  0  00  0      0 3228852   2076 594184    0    0     0     0  176  268  0  0 100  0  00  0      0 3228852   2076 594184    0    0     0     0  173  274  0  0 100  0  0
^C
┌──[root@liruilongs.github.io]-[~]
└─$

mpstat CUP子系统监控

┌──[root@liruilongs.github.io]-[~]
└─$ mpstat 1
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/23/2021      _x86_64_        (2 CPU)01:41:22 AM  CPU    %usr   %nice    %sys %iowait    %irq   %soft  %steal  %guest  %gnice   %idle
01:41:23 AM  all    0.52    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00   99.48
01:41:24 AM  all    0.00    0.00    0.51    0.00    0.00    0.00    0.00    0.00    0.00   99.49
01:41:25 AM  all    0.00    0.00    0.00    0.00    0.00    0.52    0.00    0.00    0.00   99.48
01:41:26 AM  all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
01:41:27 AM  all    0.00    0.00    0.52    0.00    0.00    0.00    0.00    0.00    0.00   99.48
01:41:28 AM  all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
01:41:29 AM  all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
^C
Average:     all    0.07    0.00    0.15    0.00    0.00    0.07    0.00    0.00    0.00   99.71
┌──[root@liruilongs.github.io]-[~]
└─$ mpstat 1 3
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/23/2021      _x86_64_        (2 CPU)01:42:06 AM  CPU    %usr   %nice    %sys %iowait    %irq   %soft  %steal  %guest  %gnice   %idle
01:42:07 AM  all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
01:42:08 AM  all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
01:42:09 AM  all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
Average:     all    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
┌──[root@liruilongs.github.io]-[~]
└─$

iostat IO子系统监控

┌──[root@liruilongs.github.io]-[~]
└─$ iostat 1 3
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/23/2021      _x86_64_        (2 CPU)avg-cpu:  %user   %nice %system %iowait  %steal   %idle0.21    0.00    0.24    0.05    0.00   99.50Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sda               0.56        19.86        12.15     387999     237243avg-cpu:  %user   %nice %system %iowait  %steal   %idle0.00    0.00    0.00    0.52    0.00   99.48Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sda               1.00         8.00         0.00          8          0avg-cpu:  %user   %nice %system %iowait  %steal   %idle0.00    0.00    0.00    0.00    0.00  100.00Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sda               0.00         0.00         0.00          0          0┌──[root@liruilongs.github.io]-[~]
└─$

top 综合情况监控

top - 01:46:43 up  5:29,  1 user,  load average: 0.01, 0.02, 0.05
Tasks: 151 total,   1 running, 150 sleeping,   0 stopped,   0 zombie
%Cpu0  :  0.0 us,  0.0 sy,  0.0 ni,100.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
%Cpu1  :  0.3 us,  0.0 sy,  0.0 ni, 99.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem :  4030172 total,  3227596 free,   205988 used,   596588 buff/cache
KiB Swap: 10485756 total, 10485756 free,        0 used.  3531292 avail MemPID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND844 etcd      20   0 10.308g  22996  11100 S   2.7  0.6  11:10.71 etcd572 root      20   0  220312   5208   3508 S   0.3  0.1   0:06.03 rsyslogd592 root      20   0  298868   6260   4916 S   0.3  0.2   1:32.71 vmtoolsd1 root      20   0  125408   3836   2440 S   0.0  0.1   0:09.03 systemd2 root      20   0       0      0      0 S   0.0  0.0   0:00.07 kthreadd3 root      20   0       0      0      0 S   0.0  0.0   0:00.61 ksoftirqd/05 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kworker/0:0H7 root      rt   0       0      0      0 S   0.0  0.0   0:00.27 migration/08 root      20   0       0      0      0 S   0.0  0.0   0:00.00 rcu_bh9 root      20   0       0      0      0 S   0.0  0.0   0:05.72 rcu_sched10 root      rt   0       0      0      0 S   0.0  0.0   0:01.25 watchdog/011 root      rt   0       0      0      0 S   0.0  0.0   0:01.15 watchdog/112 root      rt   0       0      0      0 S   0.0  0.0   0:00.17 migration/113 root      20   0       0      0      0 S   0.0  0.0   0:00.54 ksoftirqd/115 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kworker/1:0H17 root      20   0       0      0      0 S   0.0  0.0   0:00.05 kdevtmpfs18 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 netns19 root      20   0       0      0      0 S   0.0  0.0   0:00.03 khungtaskd20 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 writeback21 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kintegrityd22 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 bioset23 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kblockd

load average: 0.01 不能超过 0.7
需要除以 cpu核数；数字1

2、检测系统历史运行负载：sar

┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ pwd
/usr/lib/systemd/system
┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ cat sysstat.service
# /usr/lib/systemd/system/sysstat.service
# (C) 2012 Peter Schiffer (pschiffe <at> redhat.com)
#
# sysstat-10.1.5 systemd unit file:
#        Insert a dummy record in current daily data file.
#        This indicates that the counters have restarted from 0.[Unit]
Description=Resets System Activity Logs[Service]
Type=oneshot
RemainAfterExit=yes
User=root
ExecStart=/usr/lib64/sa/sa1 --boot[Install]
WantedBy=multi-user.target┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ systemctl status  sysstat.service
● sysstat.service - Resets System Activity LogsLoaded: loaded (/usr/lib/systemd/system/sysstat.service; enabled; vendor preset: enabled)Active: active (exited) since Wed 2021-10-13 01:53:41 CST; 1 weeks 3 days agoMain PID: 584 (code=exited, status=0/SUCCESS)CGroup: /system.slice/sysstat.serviceOct 13 01:53:41 liruilongs.github.io systemd[1]: Starting Resets System Activity Logs...
Oct 13 01:53:41 liruilongs.github.io systemd[1]: Started Resets System Activity Logs.

红帽8没有，需要自己开启，红帽7有

┌──[root@liruilongs.github.io]-[/]
└─$ cd /var/log/sa
┌──[root@liruilongs.github.io]-[/var/log/sa]
└─$ ls
sa01  sa03  sa04  sa05  sa06  sa08  sa09  sa13  sa14  sa23  sa24  sa30  sar03

CPU 监控信息

┌──[root@liruilongs.github.io]-[/var/log/sa]
└─$ sar -f sa24 -p # CPU 监控信息
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/24/2021      _x86_64_        (2 CPU)08:14:46 AM       LINUX RESTART08:20:03 AM     CPU     %user     %nice   %system   %iowait    %steal     %idle
09:40:01 AM     all      1.44      0.00      1.72      0.21      0.00     96.64
09:50:01 AM     all      0.80      0.00      1.51      0.01      0.00     97.69
10:00:01 AM     all      0.81      0.00      1.56      0.00      0.00     97.63
10:10:01 AM     all      0.74      0.00      1.40      0.02      0.00     97.83
10:20:01 AM     all      0.65      0.00      1.26      0.00      0.00     98.08
10:30:01 AM     all      0.78      0.02      1.50      0.04      0.00     97.67
10:40:01 AM     all      3.24      0.00      5.65      0.01      0.00     91.11
Average:        all      1.18      0.00      2.10      0.03      0.00     96.69

负载信息

┌──[root@liruilongs.github.io]-[/var/log/sa]
└─$ sar -f sa24 -q #  负载信息
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/24/2021      _x86_64_        (2 CPU)08:14:46 AM       LINUX RESTART08:20:03 AM   runq-sz  plist-sz   ldavg-1   ldavg-5  ldavg-15   blocked
09:40:01 AM         0       215      0.06      0.40      0.42         0
09:50:01 AM         0       215      0.41      0.21      0.28         0
10:00:01 AM         1       215      0.27      0.13      0.20         0
10:10:01 AM         1       216      0.03      0.08      0.14         0
10:20:01 AM         0       216      0.00      0.04      0.10         0
10:30:01 AM         1       217      0.00      0.03      0.06         0
10:40:01 AM         0       215      0.07      0.16      0.14         0
Average:            0       216      0.12      0.15      0.19         0

内存信息

┌──[root@liruilongs.github.io]-[/var/log/sa]
└─$ sar -f sa24 -r #  内存信息
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/24/2021      _x86_64_        (2 CPU)08:14:46 AM       LINUX RESTART08:20:03 AM kbmemfree kbmemused  %memused kbbuffers  kbcached  kbcommit   %commit  kbactive   kbinact   kbdirty
09:40:01 AM   3472808    557364     13.83      2076    266132    832840      5.74    157400    244716         8
09:50:01 AM   3470960    559212     13.88      2076    266516    832840      5.74    158004    244652         4
10:00:01 AM   3466736    563436     13.98      2076    270432    832972      5.74    158272    248436         0
10:10:01 AM   3465552    564620     14.01      2076    270856    833480      5.74    159624    247772         0
10:20:01 AM   3465732    564440     14.01      2076    271172    833584      5.74    159720    248028         0
10:30:01 AM   3457092    573080     14.22      2076    272904    833772      5.74    161872    247732         0
10:40:01 AM   3458440    571732     14.19      2076    271864    833032      5.74    160824    247256         0
Average:      3465331    564841     14.02      2076    269982    833217      5.74    159388    246942         2

网卡信息

┌──[root@liruilongs.github.io]-[/var/log/sa]
└─$ sar -f sa24 -n DEV #  网卡信息
Linux 3.10.0-693.el7.x86_64 (liruilongs.github.io)      10/24/2021      _x86_64_        (2 CPU)08:14:46 AM       LINUX RESTART08:20:03 AM     IFACE   rxpck/s   txpck/s    rxkB/s    txkB/s   rxcmp/s   txcmp/s  rxmcst/s
09:40:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
09:40:01 AM     ens32      0.34      0.37      0.04      0.33      0.00      0.00      0.00
09:40:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
09:40:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
09:50:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
09:50:01 AM     ens32      1.19      0.78      0.10      0.36      0.00      0.00      0.00
09:50:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
09:50:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:00:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:00:01 AM     ens32      0.20      0.13      0.02      0.01      0.00      0.00      0.00
10:00:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
10:00:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:10:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:10:01 AM     ens32      0.49      0.50      0.04      0.54      0.00      0.00      0.00
10:10:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
10:10:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:20:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:20:01 AM     ens32      0.17      0.11      0.01      0.01      0.00      0.00      0.00
10:20:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
10:20:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:30:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:30:01 AM     ens32      0.04      0.03      0.00      0.00      0.00      0.00      0.00
10:30:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
10:30:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:40:01 AM br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
10:40:01 AM     ens32      0.08      0.06      0.01      0.00      0.00      0.00      0.00
10:40:01 AM        lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
10:40:01 AM   docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00Average:        IFACE   rxpck/s   txpck/s    rxkB/s    txkB/s   rxcmp/s   txcmp/s  rxmcst/s
Average:    br-4b3da203747c      0.00      0.00      0.00      0.00      0.00      0.00      0.00
Average:        ens32      0.36      0.28      0.03      0.17      0.00      0.00      0.00
Average:           lo      0.13      0.13      0.01      0.01      0.00      0.00      0.00
Average:      docker0      0.00      0.00      0.00      0.00      0.00      0.00      0.00
┌──[root@liruilongs.github.io]-[/var/log/sa]
└─$

3、通过 Performance Co-Pilot(PCP) 收集性能数据

安装相关包

┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ yum -y install pcp
┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ yum install pcp-gui -y

启动相关服务

┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ systemctl enable --now pmcd # 启动监控服务
Created symlink from /etc/systemd/system/multi-user.target.wants/pmcd.service to /usr/lib/systemd/system/pmcd.service.
┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]
└─$ systemctl enable --now pmlogger # 启动存数据的数据库
Created symlink from /etc/systemd/system/multi-user.target.wants/pmlogger.service to /usr/lib/systemd/system/pmlogger.service.
┌──[root@liruilongs.github.io]-[/usr/lib/systemd/system]

通过pmchart，pminfo，pmval读取系统当前及历史运行数据

查看所有的监控内容

┌──[root@liruilongs.github.io]-[~]
└─$ pminfo
jbd2.njournals
jbd2.transaction.count
jbd2.transaction.requested
......
┌──[root@liruilongs.github.io]-[~]
└─$ pminfo | wc -l
2052

查看所有的监控参数详细信息

┌──[root@liruilongs.github.io]-[~]
└─$ pminfo -dt proc.nprocs #参数详细信息proc.nprocs [instantaneous number of processes]Data Type: 32-bit unsigned int  InDom: PM_INDOM_NULL 0xffffffffSemantics: instant  Units: none
┌──[root@liruilongs.github.io]-[~]
└─$ pminfo  proc.nprocs #进程监控
proc.nprocs

监控进程信息 pmval proc.nprocs

┌──[root@liruilongs.github.io]-[~]
└─$ pmval  proc.nprocs #进程监控metric:    proc.nprocs
host:      liruilongs.github.io
semantics: instantaneous value
units:     none
samples:   all158158158158158158
^C

监控剩余内存的量 pmval -t 2 mem.freemem

┌──[root@liruilongs.github.io]-[~]
└─$ pmval  -t 2 mem.freemem # 剩余内存信息metric:    mem.freemem
host:      liruilongs.github.io
semantics: instantaneous value
units:     Kbyte
samples:   all34716123471636347163634716363471636
^C
┌──[root@liruilongs.github.io]-[~]
└─$

查看监控日志

┌──[root@liruilongs.github.io]-[~]
└─$ pminfo | grep mem.freemem
mem.freemem
┌──[root@liruilongs.github.io]-[~]
└─$ cd /var/log/pcp
┌──[root@liruilongs.github.io]-[/var/log/pcp]
└─$ ls
NOTICES  pmcd  pmie  pmlogger  pmproxy
┌──[root@liruilongs.github.io]-[/var/log/pcp]
└─$ cd pmlogger/
┌──[root@liruilongs.github.io]-[/var/log/pcp/pmlogger]
└─$ ls
liruilongs.github.io  pmlogger_check.log.prev  pmlogger_daily-K.log.prev  pmlogger_daily.stamp
┌──[root@liruilongs.github.io]-[/var/log/pcp/pmlogger]
└─$ cd liruilongs.github.io/
┌──[root@liruilongs.github.io]-[/var/log/pcp/pmlogger/liruilongs.github.io]
└─$ ls
20211023.02.05.0.xz     20211023.02.30-00.0.xz     20211023.02.30.0.xz     20211024.08.16.0      Latest
20211023.02.05.index    20211023.02.30-00.index    20211023.02.30.index    20211024.08.16.index  pmlogger.log
20211023.02.05.meta.xz  20211023.02.30-00.meta.xz  20211023.02.30.meta.xz  20211024.08.16.meta   pmlogger.log.prior
┌──[root@liruilongs.github.io]-[/var/log/pcp/pmlogger/liruilongs.github.io]
└─$ pmval -a 20211024.08.16.0 mem.freemem  # 查看监控日志 每秒metric:    mem.freemem
archive:   20211024.08.16.0
host:      liruilongs.github.io
start:     Sun Oct 24 08:17:06 2021
end:       Sun Oct 24 10:02:06 2021
semantics: instantaneous value
units:     Kbyte
samples:   6301
interval:  1.00 sec
08:17:06.095  No values available
08:17:07.095  No values available
08:17:08.095              3651184
08:17:09.095              3651184
08:17:10.095              3651184
08:17:11.095              3651184
08:17:12.095              3651184
08:17:13.095              3651184
08:17:14.095              3651184
08:17:15.095              3651184
08:17:16.095              3651184
08:17:17.095              3651184
08:17:18.095              3651184
08:17:19.095              3651184
08:17:20.095              3651184
08:17:21.095              3651184
08:17:22.095              3651184
08:17:23.095              3651184
08:17:24.095              3651184
08:17:25.095              3651184
08:17:26.095              3651184
......

加时间粒度

┌──[root@liruilongs.github.io]-[/var/log/pcp/pmlogger/liruilongs.github.io]
└─$ pmval -a 20211024.08.16.0 -t 500 mem.freemem  # 加时间粒度metric:    mem.freemem
archive:   20211024.08.16.0
host:      liruilongs.github.io
start:     Sun Oct 24 08:17:06 2021
end:       Sun Oct 24 10:08:06 2021
semantics: instantaneous value
units:     Kbyte
samples:   14
interval:  500.00 sec
08:17:06.095  No values available
08:25:26.095              3488128
08:33:46.095              3488128
08:42:06.095              3488128
08:50:26.095              3488128
08:58:46.095              3488128
09:07:06.095              3488128
09:15:26.095              3488128
09:23:46.095              3488128
09:32:06.095              3488128
09:40:26.095              3472784
09:48:46.095              3472024
09:57:06.095              3467052
10:05:26.095              3466276
┌──[root@liruilongs.github.io]-[/var/log/pcp/pmlogger/liruilongs.github.io]
└─$

二、如何进行内核参数调优

/proc & /sys

/proc 存放内存中数据，每次重启就没了。

┌──[root@liruilongs.github.io]-[/]
└─$ cd proc/
┌──[root@liruilongs.github.io]-[/proc]
└─$ ls
1      16836  241  269  284  3    313  362  45   647   853        crypto       keys        pagetypeinfo   tty
10     16842  242  270  285  300  314  363  469  649   856        devices      key-users   partitions     uptime
1041   16843  250  271  286  301  315  364  47   66    857        diskstats    kmsg        sched_debug    version
1043   16850  251  272  287  302  316  365  5    6671  861        dma          kpagecount  schedstat      vmallocinfo
1045   16862  253  273  288  303  317  366  506  7     864        driver       kpageflags  scsi           vmstat
1050   17     254  274  289  304  318  367  509  7603  8922       execdomains  loadavg     self           zoneinfo
1051   18     255  275  290  305  319  368  570  7604  9          fb           locks       slabinfo
1054   19     256  276  291  306  32   369  581  7606  98         filesystems  mdstat      softirqs
1055   2      262  277  292  307  320  370  583  8     acpi       fs           meminfo     stat
1056   20     263  278  293  308  321  371  585  8095  buddyinfo  interrupts   misc        swaps
11     21     264  279  294  309  322  372  594  8158  bus        iomem        modules     sys
12     22     265  280  295  31   33   373  595  8159  cgroups    ioports      mounts      sysrq-trigger
13     23     266  281  296  310  34   42   596  8160  cmdline    irq          mpt         sysvipc
15     24     267  282  298  311  348  44   627  8161  consoles   kallsyms     mtrr        timer_list
16006  240    268  283  299  312  361  441  642  8162  cpuinfo    kcore        net         timer_stats
┌──[root@liruilongs.github.io]-[/proc]
└─$ pwd
/proc
┌──[root@liruilongs.github.io]-[/proc]
└─$ cd sys
┌──[root@liruilongs.github.io]-[/proc/sys]
└─$ ls
abi  crypto  debug  dev  fs  kernel  net  user  vm

内存调优目录

┌──[root@liruilongs.github.io]-[/proc/sys]
└─$ cd vm # 内存调优目录
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ ls
admin_reserve_kbytes       hugepages_treat_as_movable  mmap_min_addr             overcommit_memory
block_dump                 hugetlb_shm_group           mmap_rnd_bits             overcommit_ratio
compact_memory             laptop_mode                 mmap_rnd_compat_bits      page-cluster
dirty_background_bytes     legacy_va_layout            nr_hugepages              panic_on_oom
dirty_background_ratio     lowmem_reserve_ratio        nr_hugepages_mempolicy    percpu_pagelist_fraction
dirty_bytes                max_map_count               nr_overcommit_hugepages   stat_interval
dirty_expire_centisecs     memory_failure_early_kill   nr_pdflush_threads        swappiness
dirty_ratio                memory_failure_recovery     numa_zonelist_order       user_reserve_kbytes
dirty_writeback_centisecs  min_free_kbytes             oom_dump_tasks            vfs_cache_pressure
drop_caches                min_slab_ratio              oom_kill_allocating_task  zone_reclaim_mode
extfrag_threshold          min_unmapped_ratio          overcommit_kbytes
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$

查看内核参数：cat & sysctl -a

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ sysctl -a  # 查看所有调优参数
abi.vsyscall32 = 1
crypto.fips_enabled = 0
debug.exception-trace = 1
debug.kprobes-optimization = 1
debug.panic_on_rcu_stall = 0
dev.hpet.max-user-freq = 64
dev.mac_hid.mouse_button2_keycode = 97
dev.mac_hid.mouse_button3_keycode = 100
dev.mac_hid.mouse_button_emulation = 0
dev.parport.default.spintime = 500
dev.parport.default.timeslice = 200
dev.raid.speed_limit_max = 200000
dev.raid.speed_limit_min = 1000
dev.scsi.logging_level = 0
fs.aio-max-nr = 65536
fs.aio-nr = 0
..............

根据变量找对应文件参数

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ sysctl -a | grep net.ipv4.ip_forward
net.ipv4.ip_forward = 1
net.ipv4.ip_forward_use_pmtu = 0
sysctl: reading key "net.ipv6.conf.all.stable_secret"
sysctl: reading key "net.ipv6.conf.br-4b3da203747c.stable_secret"
sysctl: reading key "net.ipv6.conf.default.stable_secret"
sysctl: reading key "net.ipv6.conf.docker0.stable_secret"
sysctl: reading key "net.ipv6.conf.ens32.stable_secret"
sysctl: reading key "net.ipv6.conf.lo.stable_secret"
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cd ../net/ipv
ipv4/ ipv6/
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cd ../net/ipv4/
┌──[root@liruilongs.github.io]-[/proc/sys/net/ipv4]
└─$ cat ip_forward
1

调整内核参数：echo & sysctl

设置调优参数：临时调整

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cat swappiness
30
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ echo 40 > swappiness  ## 临时调整
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cat swappiness
40

设置调优参数：永久调整

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cat /etc/sysc
sysconfig/   sysctl.conf  sysctl.d/
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cat /etc/sysctl.conf
# sysctl settings are defined through files in
# /usr/lib/sysctl.d/, /run/sysctl.d/, and /etc/sysctl.d/.
#
# Vendors settings live in /usr/lib/sysctl.d/.
# To override a whole file, create a new file with the same in
# /etc/sysctl.d/ and put new settings there. To override
# only specific settings, add a file with a lexically later
# name in /etc/sysctl.d/ and put new settings there.
#
# For more information, see sysctl.conf(5) and sysctl.d(5).
net.ipv4.ip_forward = 1
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ echo "vm.swappiness = 20" >>  /etc/sysctl.conf  ## 永久调整
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cat /etc/sysctl.conf
# sysctl settings are defined through files in
# /usr/lib/sysctl.d/, /run/sysctl.d/, and /etc/sysctl.d/.
#
# Vendors settings live in /usr/lib/sysctl.d/.
# To override a whole file, create a new file with the same in
# /etc/sysctl.d/ and put new settings there. To override
# only specific settings, add a file with a lexically later
# name in /etc/sysctl.d/ and put new settings there.
#
# For more information, see sysctl.conf(5) and sysctl.d(5).
net.ipv4.ip_forward = 1
vm.swappiness = 20
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ sysctl -p
net.ipv4.ip_forward = 1
vm.swappiness = 20
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$

内核调优之清理缓存

如何使用帮助调整内核参数

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ man -K drop_caches
--Man-- next: proc(5) [ view (return) | skip (Ctrl-D) | quit (Ctrl-C) ]

清理缓存

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ cat drop_caches  #缓存处理
0
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ man -K drop_caches
No manual entry for drop_caches
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ man -K drop_caches
No manual entry for drop_caches
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ man -K ip_forward
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ free -mtotal        used        free      shared  buff/cache   available
Mem:           3935         212        3357          16         366        3440
Swap:         10239           0       10239
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ echo 3 > /proc/sys/vm/drop_caches
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ free -mtotal        used        free      shared  buff/cache   available
Mem:           3935         200        3575          16         159        3504
Swap:         10239           0       10239
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$

调整内核模块参数：lsmod、modinfo、modprobe

加载的内核模块列表 lsmod

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ lsmod               # 大小 调用次数
Module                  Size  Used by
binfmt_misc            17468  1
xt_conntrack           12760  2
ipt_MASQUERADE         12678  2
nf_nat_masquerade_ipv4    13412  1 ipt_MASQUERADE
nf_conntrack_netlink    40449  0
nfnetlink              14696  2 nf_conntrack_netlink
xt_addrtype            12676  2
iptable_filter         12810  1
iptable_nat            12875  1
nf_conntrack_ipv4      15053  3
nf_defrag_ipv4         12729  1 nf_conntrack_ipv4
。。。。。
。。。。。。
crct10dif_common       12595  3 crct10dif_pclmul,crct10dif_generic,crc_t10dif
crc32c_intel           22079  1
ahci                   34042  0
drm                   370825  2 ttm,drm_kms_helper
libahci                31992  1 ahci
mptspi                 22542  2
ata_piix               35038  0
scsi_transport_spi     30732  1 mptspi
mptscsih               40150  1 mptspi
libata                238896  5 ahci,pata_acpi,libahci,ata_generic,ata_piix
e1000                 137500  0
serio_raw              13413  0
mptbase               105960  2 mptspi,mptscsih
i2c_core               40756  3 drm,i2c_piix4,drm_kms_helper
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$

查看内核模块参数 modinfo

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ modinfo kvm # 查看内核模块参数
filename:       /lib/modules/3.10.0-693.el7.x86_64/kernel/arch/x86/kvm/kvm.ko.xz
license:        GPL
author:         Qumranet
rhelversion:    7.4
srcversion:     FA3AAB0FB1DD5C7B9D69811
depends:        irqbypass
intree:         Y
vermagic:       3.10.0-693.el7.x86_64 SMP mod_unload modversions
signer:         CentOS Linux kernel signing key
sig_key:        DA:18:7D:CA:7D:BE:53:AB:05:BD:13:BD:0C:4E:21:F4:22:B6:A4:9C
sig_hashalgo:   sha256
parm:           ignore_msrs:bool
parm:           min_timer_period_us:uint
parm:           kvmclock_periodic_sync:bool
parm:           tsc_tolerance_ppm:uint
parm:           lapic_timer_advance_ns:uint
parm:           vector_hashing:bool
parm:           halt_poll_ns:uint
parm:           halt_poll_ns_grow:uint
parm:           halt_poll_ns_shrink:uint
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$

内核模块调优之嵌套虚拟化设置

┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ modinfo kvm
filename:       /lib/modules/3.10.0-693.el7.x86_64/kernel/arch/x86/kvm/kvm.ko.xz
license:        GPL
author:         Qumranet
rhelversion:    7.4
srcversion:     FA3AAB0FB1DD5C7B9D69811
depends:        irqbypass
intree:         Y
vermagic:       3.10.0-693.el7.x86_64 SMP mod_unload modversions
signer:         CentOS Linux kernel signing key
sig_key:        DA:18:7D:CA:7D:BE:53:AB:05:BD:13:BD:0C:4E:21:F4:22:B6:A4:9C
sig_hashalgo:   sha256
parm:           ignore_msrs:bool
parm:           min_timer_period_us:uint
parm:           kvmclock_periodic_sync:bool
parm:           tsc_tolerance_ppm:uint
parm:           lapic_timer_advance_ns:uint
parm:           vector_hashing:bool
parm:           halt_poll_ns:uint
parm:           halt_poll_ns_grow:uint
parm:           halt_poll_ns_shrink:uint
┌──[root@liruilongs.github.io]-[/proc/sys/vm]
└─$ modinfo kvm | grep  ignore_msrs # 设置允许嵌套虚拟化
parm:           ignore_msrs:bool
┌──[root@liruilongs.github.io]-[/etc/modprobe.d]
└─$ echo "options kvm ignore_msrs=1" >> /etc/modprobe.d/kvm.conf  ##从起机器

如何限制服务的资源占用

资源消耗分为用户和服务，消耗：CPU，内存和 I/O

经典的限制通过 `pam_limit` 模块配合 `/etc/security/limits.conf` 实现（用户）

用户登录会加载pam_limit 模块，pam_limit 模块读取配置文件 /etc/security/limits.conf限制用户资源的占用

┌──[root@liruilongs.github.io]-[/etc/modprobe.d]
└─$ cat /etc/security/limits.conf
# /etc/security/limits.conf
#
#This file sets the resource limits for the users logged in via PAM.
#It does not affect resource limits of the system services.
#
#Also note that configuration files in /etc/security/limits.d directory,
#which are read in alphabetical order, override the settings in this
#file in case the domain is the same or more specific.
#That means for example that setting a limit for wildcard domain here
#can be overriden with a wildcard setting in a config file in the
#subdirectory, but a user specific setting here can be overriden only
#with a user specific setting in the subdirectory.
#
#Each line describes a limit for a user in the form:
#
#<domain>        <type>  <item>  <value>
#
#Where:
#<domain> can be:
#        - a user name
#        - a group name, with @group syntax
#        - the wildcard *, for default entry
#        - the wildcard %, can be also used with %group syntax,
#                 for maxlogin limit
#
#<type> can have the two values:
#        - "soft" for enforcing the soft limits
#        - "hard" for enforcing hard limits
#
#<item> can be one of the following:
#        - core - limits the core file size (KB) 核心文件大小
#        - data - max data size (KB)
#        - fsize - maximum filesize (KB)
#        - memlock - max locked-in-memory address space (KB)
#        - nofile - max number of open file descriptors # 访问文件数量
#        - rss - max resident set size (KB)
#        - stack - max stack size (KB)
#        - cpu - max CPU time (MIN)  # cup时间
#        - nproc - max number of processes
#        - as - address space limit (KB)
#        - maxlogins - max number of logins for this user #最多的登录数
#        - maxsyslogins - max number of logins on the system
#        - priority - the priority to run user process with
#        - locks - max number of file locks the user can hold
#        - sigpending - max number of pending signals
#        - msgqueue - max memory used by POSIX message queues (bytes)
#        - nice - max nice priority allowed to raise to values: [-20, 19]
#        - rtprio - max realtime priority
#
#<domain>      <type>  <item>         <value>
##*               soft    core            0
#*               hard    rss             10000
#@student        hard    nproc           20
#@faculty        soft    nproc           20
#@faculty        hard    nproc           50
#ftp             hard    nproc           0
#@student        -       maxlogins       4# End of file

限制服务的资源之限制用户通过ssh的登录数

┌──[root@liruilongs.github.io]-[/etc/modprobe.d]
└─$vim /etc/security/limits.conf
student       hard       maxlogins       4

也可以通过 cgroup 实现资源的限制(服务)

┌──[root@liruilongs.github.io]-[/etc/modprobe.d]
└─$ md5sum /dev/urandom & #启动一个进程 cpu100%
[1] 38162
┌──[root@liruilongs.github.io]-[/etc/modprobe.d]
└─$ top
top - 13:37:42 up  4:10,  1 user,  load average: 1.03, 0.80, 0.43
Tasks: 158 total,   2 running, 156 sleeping,   0 stopped,   0 zombie
%Cpu(s):  2.0 us, 53.9 sy,  0.0 ni, 43.3 id,  0.0 wa,  0.0 hi,  0.8 si,  0.0 st
KiB Mem :  4030172 total,  3637840 free,   207252 used,   185080 buff/cache
KiB Swap: 10485756 total, 10485756 free,        0 used.  3575580 avail MemPID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND861 etcd      20   0 10.254g  22572  11092 S   9.7  0.6  22:07.49 etcd9 root      20   0       0      0      0 S   1.7  0.0   0:28.00 rcu_sched595 root      20   0  298868   6260   4916 S   1.6  0.2   2:27.63 vmtoolsd371 root      20   0       0      0      0 S   0.4  0.0   1:23.11 xfsaild/sda11 root      20   0   51688   3960   2624 S   0.4  0.1   0:31.08 systemd36765 root      20   0       0      0      0 S   0.2  0.0   0:03.13 kworker/1:0583 root      20   0   21616   1284    972 S   0.2  0.0   0:09.59 irqbalance857 root      20   0  573468  16760   6020 S   0.2  0.4   0:14.24 tuned13 root      20   0       0      0      0 S   0.2  0.0   0:06.16 ksoftirqd/1864 root      20   0 1015116  42652  13740 S   0.1  1.1   0:22.33 containerd853 root      20   0  228220   5128   3240 S   0.1  0.1   0:09.13 httpd594 root      20   0  224504  12424   3348 S   0.1  0.3   0:08.61 rsyslogd649 chrony    20   0  117704   1772   1312 S   0.1  0.0   0:03.47 chronyd7604 root      20   0  151984   5352   4088 R   0.1  0.1   0:05.89 sshd38166 root      20   0  161908   2264   1556 R   0.1  0.1   0:00.10 top1055 root      20   0   81744   3132   2168 S   0.0  0.1   0:10.09 pmdalinux8095 root      20   0       0      0      0 S   0.0  0.0   0:01.54 kworker/u256:011 root      rt   0       0      0      0 S   0.0  0.0   0:02.17 watchdog/11051 root      20   0   81628   2700   1912 S   0.0  0.1   0:03.07 pmdaproc2 root      20   0       0      0      0 S   0.0  0.0   0:00.09 kthreadd3 root      20   0       0      0      0 S   0.0  0.0   0:01.56 ksoftirqd/05 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kworker/0:0H7 root      rt   0       0      0      0 S   0.0  0.0   0:01.08 migration/0
[1]+  Terminated              md5sum /dev/urandom
┌──[root@liruilongs.github.io]-[/etc/modprobe.d]
└─$

现在我们把这个写成一个服务，然后通过cgroup做简单限制。

┌──[root@liruilongs.github.io]-[~]
└─$ vim /etc/systemd/system/md5sum.service
┌──[root@liruilongs.github.io]-[~]
└─$ cat /etc/systemd/system/md5sum.service
[Unit]
Description=MD5 Demo[Service]
CPUQuota=40%
ExecStart=/usr/bin/md5sum /dev/urandom[Install]
WantedBy=multi-user.target
┌──[root@liruilongs.github.io]-[~]
└─$
┌──[root@liruilongs.github.io]-[~]
└─$ systemctl daemon-reload
┌──[root@liruilongs.github.io]-[~]
└─$ systemctl start md5sum.service
┌──[root@liruilongs.github.io]-[~]
└─$ top
top - 19:12:14 up  9:44,  1 user,  load average: 0.29, 0.20, 0.11
Tasks: 159 total,   2 running, 157 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.5 us, 20.3 sy,  0.0 ni, 78.8 id,  0.0 wa,  0.0 hi,  0.4 si,  0.0 st
KiB Mem :  4030172 total,  3606960 free,   209832 used,   213380 buff/cache
KiB Swap: 10485756 total, 10485756 free,        0 used.  3558764 avail MemPID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND59598 root      20   0  107920    612    516 R  39.7  0.0   0:02.54 md5sum  ##CPU限制为40%861 etcd      20   0 10.254g  22572  11092 S   6.3  0.6  51:24.88 etcd59601 root      20   0  161908   2268   1560 R   1.0  0.1   0:00.08 top595 root      20   0  298868   6260   4916 S   0.7  0.2   5:35.80 vmtoolsd1 root      20   0   51688   4000   2648 S   0.3  0.1   1:15.99 systemd9 root      20   0       0      0      0 S   0.3  0.0   0:50.90 rcu_sched371 root      20   0       0      0      0 S   0.3  0.0   3:02.41 xfsaild/sda16671 root      20   0 1309148  66444  25620 S   0.3  1.6   0:48.35 dockerd2 root      20   0       0      0      0 S   0.0  0.0   0:00.13 kthreadd3 root      20   0       0      0      0 S   0.0  0.0   0:04.25 ksoftirqd/05 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kworker/0:0H7 root      rt   0       0      0      0 S   0.0  0.0   0:02.39 migration/08 root      20   0       0      0      0 S   0.0  0.0   0:00.00 rcu_bh10 root      rt   0       0      0      0 S   0.0  0.0   0:03.13 watchdog/011 root      rt   0       0      0      0 S   0.0  0.0   0:04.80 watchdog/112 root      rt   0       0      0      0 S   0.0  0.0   0:02.72 migration/113 root      20   0       0      0      0 S   0.0  0.0   0:11.18 ksoftirqd/115 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kworker/1:0H17 root      20   0       0      0      0 S   0.0  0.0   0:00.11 kdevtmpfs18 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 netns19 root      20   0       0      0      0 S   0.0  0.0   0:00.08 khungtaskd20 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 writeback21 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kintegrityd
┌──[root@liruilongs.github.io]-[~]
└─$

这个具体小伙伴可以看看这篇博客: https://blog.csdn.net/sanhewuyang/article/details/120735766

关于cgroup参数设置可以使用帮助文档

┌──[root@liruilongs.github.io]-[~]
└─$ man -k systemd
┌──[root@liruilongs.github.io]-[~]
└─$ man systemd.resource-control
┌──[root@liruilongs.github.io]-[~]
└─$

为 systemd unit 实现资源限制

这个老师没讲，以后研究下，应该也和cgroup有关

自定义tuned调优配置集：

tuned 实现系统傻瓜化和集成化调优,操作系统预装多个调优场景配置集,这是一道RHCAS的考试题

查看调优策略

┌──[root@liruilongs.github.io]-[~]
└─$ tuned-adm list
Available profiles:
- balanced                    - General non-specialized tuned profile
- desktop                     - Optimize for the desktop use-case
- latency-performance         - Optimize for deterministic performance at the cost of increased power consumption
- network-latency             - Optimize for deterministic performance at the cost of increased power consumption, focused on low latency network performance
- network-throughput          - Optimize for streaming network throughput, generally only necessary on older CPUs or 40G+ networks
- powersave                   - Optimize for low power consumption
- throughput-performance      - Broadly applicable tuning that provides excellent performance across a variety of common server workloads
- virtual-guest               - Optimize for running inside a virtual guest
- virtual-host                - Optimize for running KVM guests
Current active profile: virtual-guest
┌──[root@liruilongs.github.io]-[~]
└─$ tuned-adm recommend  # 查看推荐的策略
virtual-guest
┌──[root@liruilongs.github.io]-[~]
└─$ tuned-adm profile virtual-guest

调优策略位置

┌──[root@liruilongs.github.io]-[~]
└─$ cd /usr/lib/tuned/;ls ## 调优参数
balanced   latency-performance  powersave               virtual-guest
desktop    network-latency      recommend.conf          virtual-host
functions  network-throughput   throughput-performance
┌──[root@liruilongs.github.io]-[/usr/lib/tuned]
└─$ cd virtual-guest/
┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$ ls
tuned.conf
┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$ cat tuned.conf
#
# tuned configuration
#[main]
summary=Optimize for running inside a virtual guest
include=throughput-performance[sysctl]
# If a workload mostly uses anonymous memory and it hits this limit, the entire
# working set is buffered for I/O, and any more write buffering would require
# swapping, so it's time to throttle writes until I/O can catch up.  Workloads
# that mostly use file mappings may be able to use even higher values.
#
# The generator of dirty data starts writeback at this percentage (system default
# is 20%)
vm.dirty_ratio = 30# Filesystem I/O is usually much more efficient than swapping, so try to keep
# swapping low.  It's usually safe to go even lower than this on systems with
# server-grade storage.
vm.swappiness = 30
┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$

在/etc/tuned目录中编辑自定义tuned profile,配置集中的条目可以相互引用，并通过各种plugin自定义
获取 yum仓库的tuned profile

┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$ yum list | grep tuned
tuned.noarch                             2.8.0-5.el7                   @anaconda
tuned.noarch                             2.11.0-11.el7_9               updates
tuned-gtk.noarch                         2.11.0-11.el7_9               updates
tuned-profiles-atomic.noarch             2.11.0-11.el7_9               updates
tuned-profiles-compat.noarch             2.11.0-11.el7_9               updates
tuned-profiles-cpu-partitioning.noarch   2.11.0-11.el7_9               updates
tuned-profiles-mssql.noarch              2.11.0-11.el7_9               updates
tuned-profiles-oracle.noarch             2.11.0-11.el7_9               updates
tuned-utils.noarch                       2.11.0-11.el7_9               updates
tuned-utils-systemtap.noarch             2.11.0-11.el7_9               updates
┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$ yum -y install tuned-profiles-oracle.noarch
....
┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$ tuned-adm list
Available profiles:
- balanced                    - General non-specialized tuned profile
- desktop                     - Optimize for the desktop use-case
- hpc-compute                 - Optimize for HPC compute workloads
- latency-performance         - Optimize for deterministic performance at the cost of increased power consumption
- network-latency             - Optimize for deterministic performance at the cost of increased power consumption, focused on low latency network performance
- network-throughput          - Optimize for streaming network throughput, generally only necessary on older CPUs or 40G+ networks
- oracle                      - Optimize for Oracle RDBMS
- powersave                   - Optimize for low power consumption
- throughput-performance      - Broadly applicable tuning that provides excellent performance across a variety of common server workloads
- virtual-guest               - Optimize for running inside a virtual guest
- virtual-host                - Optimize for running KVM guests
Current active profile: virtual-guest
┌──[root@liruilongs.github.io]-[/usr/lib/tuned/virtual-guest]
└─$

通过SystemTap配置内核模块进行系统底层分析：

嗯，这部分有些复杂，有些包需要订阅，先记录下，以后有机会学习

通过SystemTap配置内核模块进行系统底层分析：
SystemTap 可以简易的探索测量 kernel 中任意部件
Kernel开发者通过kprobe在kernel功能前后加入测试代码
SystemTap需要gcc，kernel-debuginfo,kernel-devel软件支持
stap命令将*.stp脚本文件编译为kernel模块，通过staprun命令运行

关于 Linux中系统调优的一些笔记相关推荐

Linux系统调优详解（十二）——IO调优之磁盘测速
今天继续给大家介绍Linux运维相关知识,本文主要内容是IO调优,测试对磁盘的速度. 在进行Linux系统调优的时候,我们需要对当前系统的硬盘读写速度进行测试.此时,我们可以选择使用hdparm命令或 ...
Linux云计算架构-系统调优【CPU、IO、网络、内核参数调优】
文章目录 Linux云计算架构-系统调优[CPU.IO.网络.内核参数调优] 1. CPU调优 2. IO调优 3. 网络调优 3.1 网络冗余(主备模式) 3.2 负载均衡模式 4. 内核参数调优 ...
51CTO学习笔记--Linux运维故障排查思路与系统调优技巧视频课程（高俊峰）
51CTO学习笔记--Linux运维故障排查思路与系统调优技巧视频课程第一课 Linux运维经验分享与思路 1.一般把主机名,写到hosts下 127.0.0.1 hostname,因为 ...
linux @webserviceclient 访问超时_Linux系统调优
linux系统linux系统安装完毕后,首先要做的就是系统调优,这样会提高系统的使用效率,接下来为大家介绍一下linux系统调优方法.linuxlinux系统安装完毕后,首先要做的就是系统调优,这样会 ...
Linux 系统调优相关工具
内容: 11.1 系统调优概述 11.2 查看 CPU 负载相关工具,找出系统中使用 CPU 最多的进程 11.3 查看 Memory 运行状态相关工具,找出系统中使用内存最多的进程 11.4 查看 ...
linux内核 cpu io等待,linux系统调优之IO、cpu
首先先要再次强调一个概念:系统调优的目的不是怎样去吧那一部分做的特别优化.而是整体的全局的考虑.性能优化是为了找到系统的瓶颈并且想办法提高瓶颈最好能够去除.(个人认为万事皆有利弊,在计算机的世界里同样 ...
alsa 测试 linux_Linux低延迟服务器系统调优
最近做了一些系统和网络调优相关的测试,达到了期望的效果,有些感悟.同时,我也发现知乎上对Linux服务器低延迟技术的讨论比较欠缺(满嘴高并发现象):或者对现今cpu + 网卡的低延迟潜力认识不足(动辄 ...
问君能有几多愁，恰似不懂Linux SQL如何调优——聊聊SQL Server on Linux最佳实践
问君能有几多愁,恰似不懂Linux SQL如何调优--聊聊SQL Server on Linux最佳实践自从微软开始拥抱Linux, SQL Server 很快就推出了 Linux版本, linux ...

关于 Linux中系统调优的一些笔记

写在前面

零、系统调优原理概述：

分析系统性能的方法：

执行系统变更的步骤：

一、如何检测系统的性能瓶颈

1、检测系统当前运行负载：vmstat、mpstat、iostat、top 等

vmstat 内存子系统监控

mpstat CUP子系统监控

iostat IO子系统监控

top 综合情况监控

2、检测系统历史运行负载：sar

CPU 监控信息

负载信息

内存信息

网卡信息

3、通过 Performance Co-Pilot(PCP) 收集性能数据

安装相关包

启动相关服务

通过pmchart，pminfo，pmval读取系统当前及历史运行数据

二、如何进行内核参数调优

/proc & /sys

查看内核参数：cat & sysctl -a

调整内核参数：echo & sysctl

内核调优之清理缓存

调整内核模块参数：lsmod、modinfo、modprobe

内核模块调优之嵌套虚拟化设置

如何限制服务的资源占用

经典的限制通过 `pam_limit` 模块配合 `/etc/security/limits.conf` 实现（用户）

限制服务的资源之限制用户通过ssh的登录数

也可以通过 cgroup 实现资源的限制(服务)

为 systemd unit 实现资源限制

自定义tuned调优配置集：

通过SystemTap配置内核模块进行系统底层分析：

关于 Linux中系统调优的一些笔记相关推荐

最新文章

热门文章

关于 Linux中系统调优的一些笔记

写在前面

零、系统调优原理概述：

分析系统性能的方法：

执行系统变更的步骤：

一、如何检测系统的性能瓶颈

1、检测系统当前运行负载：vmstat、mpstat、iostat、top 等

vmstat 内存子系统监控

mpstat CUP子系统监控

iostat IO子系统监控

top 综合情况监控

2、检测系统历史运行负载：sar

CPU 监控信息

负载信息

内存信息

网卡信息

3、通过 Performance Co-Pilot(PCP) 收集性能数据

安装相关包

启动相关服务

通过pmchart，pminfo，pmval读取系统当前及历史运行数据

二、如何进行内核参数调优

/proc & /sys

查看内核参数：cat & sysctl -a

调整内核参数：echo & sysctl

内核调优之清理缓存

调整内核模块参数：lsmod、modinfo、modprobe

内核模块调优之嵌套虚拟化设置

如何限制服务的资源占用

经典的限制通过 pam_limit 模块配合 /etc/security/limits.conf 实现（用户）

限制服务的资源之限制用户通过ssh的登录数

也可以通过 cgroup 实现资源的限制(服务)

为 systemd unit 实现资源限制

自定义tuned调优配置集：

通过SystemTap配置内核模块进行系统底层分析：

关于 Linux中系统调优的一些笔记相关推荐

最新文章

热门文章

经典的限制通过 `pam_limit` 模块配合 `/etc/security/limits.conf` 实现（用户）