性能工具之 stress 工具使用教程(带源码说明)
- 2021 年 12 月 20 日
本文字数:8508 字
阅读完需:约 28 分钟
stress 是一个在 linux 下的压力测试小工具。
我看到有些人用这个工具来描述一些资源耗尽的场景,也有人用它来做混沌测试中。请使用者要注意,这个工具并不是模拟业务问题的,是模拟系统级问题的。所以用它来模拟的时候,和业务真实场景差别还是很大的。
因为在性能工作中,经常有人因为不理解工具而误用工具,所以这里我把这个工具拿出来说明一下。
一、安装 stress
yum install -y stress
二、stress 参数
[root@7DGroupT1 ~]# stress
`stress' imposes certain types of compute stress on your system
Usage: stress [OPTION [ARG]] ...
-?, --help show this help statement
--version show version statement
-v, --verbose be verbose
-q, --quiet be quiet
-n, --dry-run show what would have been done
-t, --timeout N timeout after N seconds
--backoff N wait factor of N microseconds before work starts
-c, --cpu N spawn N workers spinning on sqrt()
-i, --io N spawn N workers spinning on sync()
-m, --vm N spawn N workers spinning on malloc()/free()
--vm-bytes B malloc B bytes per vm worker (default is 256MB)
--vm-stride B touch a byte every B bytes (default is 4096)
--vm-hang N sleep N secs before free (default none, 0 is inf)
--vm-keep redirty memory instead of freeing and reallocating
-d, --hdd N spawn N workers spinning on write()/unlink()
--hdd-bytes B write B bytes per hdd worker (default is 1GB)
Example: stress --cpu 8 --io 4 --vm 2 --vm-bytes 128M --timeout 10s
Note: Numbers may be suffixed with s,m,h,d,y (time) or B,K,M,G (size).
参数可以说非常简单了。大概看一眼就可以知道,它可以模拟 CPU、IO、内存、磁盘这些常见又重要的资源消耗。下面就一一来看一下。
三、模拟 CPU
[root@7DGroupT1 ~]# stress -c 4 -t 100
top - 10:48:11 up 63 days, 23:57, 2 users, load average: 0.67, 1.41, 4.21
Tasks: 122 total, 5 running, 117 sleeping, 0 stopped, 0 zombie
%Cpu0 : 99.7 us, 0.3 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu1 :100.0 us, 0.0 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu2 :100.0 us, 0.0 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu3 :100.0 us, 0.0 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 8010528 total, 5550792 free, 1866688 used, 593048 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 5762564 avail Mem
模拟 CPU 的参数很简洁。打印一下栈看一下:
[root@s6 ~]# pstack 29253
#0 0x00007f123634761b in random () from /usr/lib64/libc.so.6
#1 0x00007f1236347b39 in rand () from /usr/lib64/libc.so.6
#2 0x0000557e9ea32dbd in hogcpu ()
#3 0x0000557e9ea3180a in main ()
[root@s6 ~]
其实代码很简单,就是一个 hogcpu 函数。源码是这样的:
int
hogcpu (void)
{
while (1)
sqrt (rand ());
return 0;
}
是不是看了之后觉得自己都能写一个了?不就是一个 while 吗?
四、模拟内存
[root@7DGroupT1 ~]# stress --vm 30 --vm-bytes 1G --vm-hang 50 --timeout 50s
[root@7DGroupT1 ~]# vmstat 1
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 0 5606796 6828 534736 0 0 4548 212 457 710 0 0 99 1 0
0 0 0 5597852 6976 544360 0 0 9712 52 666 1163 0 0 99 1 0
0 0 0 5595060 7136 545828 0 0 1752 0 440 580 0 0 98 2 0
30 0 0 2125872 7136 546040 0 0 8 0 1098 522 0 21 79 0 0
0 14 0 100896 200 211224 0 0 529748 2932 25058 43164 1 51 4 44 0
[root@7DGroupT1 ~]# sar -B 1
Linux 3.10.0-514.21.1.el7.x86_64 (7DGroupT1) 10/03/2019 _x86_64_ (4 CPU)
10:52:49 AM pgpgin/s pgpgout/s fault/s majflt/s pgfree/s pgscank/s pgscand/s pgsteal/s %vmeff
10:52:50 AM 8.00 68.00 122.00 0.00 1143.00 0.00 0.00 0.00 0.00
10:52:51 AM 0.00 0.00 29.00 0.00 25.00 0.00 0.00 0.00 0.00
10:52:52 AM 0.00 0.00 184.00 0.00 45.00 0.00 0.00 0.00 0.00
10:52:53 AM 0.00 0.00 2482.00 0.00 804.00 0.00 0.00 0.00 0.00
10:52:54 AM 870.77 2436.92 172704.62 2.31 92710.00 38558.46 26820.00 61888.46 94.66
10:52:56 AM 76853.61 618.56 34391.24 82.47 1297422.16 205238.14 404672.68 14717.53 2.41
10:52:57 AM 125560.00 300.00 4875.00 110.00 5040.00 0.00 0.00 0.00 0.00
10:52:58 AM 111080.00 0.00 8940.00 68.00 4723.00 0.00 0.00 0.00 0.00
10:52:59 AM 80944.00 0.00 5725.00 40.00 1636.00 0.00 0.00 0.00 0.00
10:53:00 AM 26224.00 300.00 37293.00 2.00 534.00 0.00 0.00 0.00 0.00
10:53:01 AM 7344.00 180.00 1092.00 0.00 17475.00 0.00 0.00 0.00 0.00
10:53:02 AM 24576.00 224.00 5725.00 41.00 1866.00 0.00 0.00 0.00 0.00
[root@7DGroupT1 ~]# sar -r 1
Linux 3.10.0-514.21.1.el7.x86_64 (7DGroupT1) 10/03/2019 _x86_64_ (4 CPU)
10:56:55 AM kbmemfree kbmemused %memused kbbuffers kbcached kbcommit %commit kbactive kbinact kbdirty
10:57:13 AM 5582520 2428008 30.31 7648 447240 5643636 70.45 1780508 443288 3400
10:57:14 AM 101544 7908984 98.73 6736 375896 37108100 463.24 7328268 367028 3404
从上面的数据来看,确实产生了很大的 page faults,这也是模拟内存消耗的过程中必然会出现的现象。之前我也强调过,看内存够不够,就是要看这个 faluts。在 Stress 中,怎么模拟的内存呢。来看一下。
int hogvm (long long bytes, long long stride, long long hang, int keep)
{
long long i;
char *ptr = 0;
char c;
int do_malloc = 1;
while (1)
{
if (do_malloc)
{
dbg (stdout, "allocating %lli bytes ...\n", bytes);
if (!(ptr = (char *) malloc (bytes * sizeof (char))))
{
err (stderr, "hogvm malloc failed: %s\n", strerror (errno));
return 1;
}
if (keep)
do_malloc = 0;
}
dbg (stdout, "touching bytes in strides of %lli bytes ...\n", stride);
for (i = 0; i < bytes; i += stride)
ptr[i] = 'Z'; /* Ensure that COW happens. */
if (hang == 0)
{
dbg (stdout, "sleeping forever with allocated memory\n");
while (1)
sleep (1024);
}
else if (hang > 0)
{
dbg (stdout, "sleeping for %llis with allocated memory\n", hang);
sleep (hang);
}
for (i = 0; i < bytes; i += stride)
{
c = ptr[i];
if (c != 'Z')
{
err (stderr, "memory corruption at: %p\n", ptr + i);
return 1;
}
}
if (do_malloc)
{
free (ptr);
dbg (stdout, "freed %lli bytes\n", bytes);
}
}
return 0;
}
就是一个死循环加上一个内存 malloc。
五、模拟磁盘
[root@7DGroupT1 ~]# stress --hdd 5 --hdd-bytes 1G
[root@7DGroupT1 ~]# top
top - 10:35:15 up 63 days, 23:44, 2 users, load average: 9.14, 8.49, 8.29
Tasks: 124 total, 2 running, 122 sleeping, 0 stopped, 0 zombie
%Cpu0 : 0.0 us, 5.8 sy, 0.0 ni, 0.0 id, 94.2 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu1 : 1.0 us, 1.0 sy, 0.0 ni, 14.4 id, 83.6 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu2 : 0.0 us, 4.4 sy, 0.0 ni, 0.0 id, 95.6 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu3 : 0.0 us, 4.1 sy, 0.0 ni, 0.0 id, 95.9 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 8010528 total, 1940088 free, 1891792 used, 4178648 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 5687416 avail Mem
[root@7DGroupT1 ~]# vmstat 1
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
1 7 0 1474132 146392 4499176 0 0 0 12 0 1 0 0 100 0 0
0 7 0 1384720 146392 4589204 0 0 0 107624 815 966 0 3 8 88 0
0 7 0 1292432 146392 4681232 0 0 0 98920 1036 987 0 3 13 84 0
0 7 0 1194932 146392 4777968 0 0 0 115344 1033 1207 0 4 0 96 0
1 6 0 1094312 146392 4880044 0 0 0 105260 928 930 0 3 5 92 0
1 6 0 998756 146416 4974944 0 0 0 102812 862 928 0 3 0 97 0
0 7 0 897960 146448 5075492 0 0 0 131200 1268 1565 1 4 5 91 0
3 4 0 1626628 146472 4347076 0 0 0 82804 1444 1206 0 8 10 81 0
0 7 0 2354208 146656 3620344 0 0 0 118112 2229 1256 0 27 2 70 0
0 7 0 3264136 146804 2709776 0 0 0 110632 1761 1506 0 9 0 90 0
0 7 0 3143120 146940 2831728 0 0 4 106896 1211 1112 0 5 0 95 0
0 7 0 2961456 146940 3016064 0 0 0 91456 1484 1298 0 6 0 94 0
^C
[root@7DGroupT1 ~]# sar -d 1
10:54:19 AM DEV tps rd_sec/s wr_sec/s avgrq-sz avgqu-sz await svctm %util
10:54:20 AM dev253-0 307.00 24.00 229976.00 749.19 126.47 454.24 3.26 100.00
10:54:20 AM dev11-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:54:20 AM DEV tps rd_sec/s wr_sec/s avgrq-sz avgqu-sz await svctm %util
10:54:21 AM dev253-0 369.00 48.00 228528.00 619.45 127.11 347.24 2.71 100.00
10:54:21 AM dev11-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:54:21 AM DEV tps rd_sec/s wr_sec/s avgrq-sz avgqu-sz await svctm %util
10:54:22 AM dev253-0 274.00 24.00 203760.00 743.74 127.45 404.61 3.65 100.00
10:54:22 AM dev11-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:54:22 AM DEV tps rd_sec/s wr_sec/s avgrq-sz avgqu-sz await svctm %util
10:54:23 AM dev253-0 262.00 0.00 202000.00 770.99 127.61 486.35 3.82 100.10
10:54:23 AM dev11-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10:54:23 AM DEV tps rd_sec/s wr_sec/s avgrq-sz avgqu-sz await svctm %util
10:54:24 AM dev253-0 288.00 0.00 232352.00 806.78 127.92 479.62 3.47 100.00
10:54:24 AM dev11-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
[root@7DGroupT1 ~]# iostat -x -d 1
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
vda 0.00 8603.00 4.00 255.00 16.00 101840.00 786.53 125.75 471.71 3.00 479.06 3.86 100.00
scd0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
vda 0.00 8952.00 4.00 294.00 16.00 117176.00 786.52 127.03 470.79 3.75 477.14 3.36 100.00
scd0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
vda 0.00 9059.00 1.00 266.00 4.00 101900.00 763.33 126.55 433.52 0.00 435.15 3.75 100.00
scd0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
vda 0.00 6234.00 0.00 273.00 0.00 96784.00 709.04 127.94 391.18 0.00 391.18 3.67 100.10
scd0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
[root@7DGroupT1 ~]# iotop
Total DISK READ : 0.00 B/s | Total DISK WRITE : 102.23 M/s
Actual DISK READ: 0.00 B/s | Actual DISK WRITE: 75.93 M/s
TID PRIO USER DISK READ DISK WRITE SWAPIN IO> COMMAND
8681 be/4 root 0.00 B/s 20.93 M/s 0.00 % 98.97 % stress --hdd 5 --hdd-bytes 1G
8677 be/4 root 0.00 B/s 20.23 M/s 0.00 % 95.92 % stress --hdd 5 --hdd-bytes 1G
8680 be/4 root 0.00 B/s 20.23 M/s 0.00 % 95.59 % stress --hdd 5 --hdd-bytes 1G
8679 be/4 root 0.00 B/s 20.61 M/s 0.00 % 95.17 % stress --hdd 5 --hdd-bytes 1G
8678 be/4 root 0.00 B/s 20.23 M/s 0.00 % 95.16 % stress --hdd 5 --hdd-bytes 1G
7298 be/4 root 0.00 B/s 0.00 B/s 0.00 % 86.91 % [kworker/u8:1]
285 be/3 root 0.00 B/s 0.00 B/s 0.00 % 86.72 % [jbd2/vda1-8]
16384 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % YDService
1 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % systemd --syst~deserialize 21
2 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [kthreadd]
3 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [ksoftirqd/0]
516 be/4 libstora 0.00 B/s 0.00 B/s 0.00 % 0.00 % lsmd -d
5 be/0 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [kworker/0:0H]
518 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % acpid
7 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [migration/0]
8 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [rcu_bh]
9 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [rcu_sched]
10 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [watchdog/0]
11 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [watchdog/1]
12 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [migration/1]
13 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [ksoftirqd/1]
526 be/4 polkitd 0.00 B/s 0.00 B/s 0.00 % 0.00 % polkitd --no-d~ Sour~ Thread]
15 be/0 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [kworker/1:0H]
16 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [watchdog/2]
17 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [migration/2]
18 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [ksoftirqd/2]
515 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % python -Es /us~in/tuned -l -P
20 be/0 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [kworker/2:0H]
21 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [watchdog/3]
22 rt/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [migration/3]
23 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [ksoftirqd/3]
25 be/0 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [kworker/3:0H]
27 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [kdevtmpfs]
28 be/0 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [netns]
29 be/4 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [khungtaskd]
30 be/0 root 0.00 B/s 0.00 B/s 0.00 % 0.00 % [writeback]
模拟磁盘,看起来效果也是非常不错的哦。来翻一下源码。
int hoghdd (long long bytes)
{
long long i, j;
int fd;
int chunk = (1024 * 1024) - 1; /* Minimize slow writing. */
char buff[chunk];
/* Initialize buffer with some random ASCII data. */
dbg (stdout, "seeding %d byte buffer with random data\n", chunk);
for (i = 0; i < chunk - 1; i++)
{
j = rand ();
j = (j < 0) ? -j : j;
j %= 95;
j += 32;
buff[i] = j;
}
buff[i] = '\n';
while (1)
{
char name[] = "./stress.XXXXXX";
if ((fd = mkstemp (name)) == -1)
{
err (stderr, "mkstemp failed: %s\n", strerror (errno));
return 1;
}
dbg (stdout, "opened %s for writing %lli bytes\n", name, bytes);
dbg (stdout, "unlinking %s\n", name);
if (unlink (name) == -1)
{
err (stderr, "unlink of %s failed: %s\n", name, strerror (errno));
return 1;
}
dbg (stdout, "fast writing to %s\n", name);
for (j = 0; bytes == 0 || j + chunk < bytes; j += chunk)
{
if (write (fd, buff, chunk) == -1)
{
err (stderr, "write failed: %s\n", strerror (errno));
return 1;
}
}
dbg (stdout, "slow writing to %s\n", name);
for (; bytes == 0 || j < bytes - 1; j++)
{
if (write (fd, &buff[j % chunk], 1) == -1)
{
err (stderr, "write failed: %s\n", strerror (errno));
return 1;
}
}
if (write (fd, "\n", 1) == -1)
{
err (stderr, "write failed: %s\n", strerror (errno));
return 1;
}
++j;
dbg (stdout, "closing %s after %lli bytes\n", name, j);
close (fd);
}
return 0;
}
死循环加上 for 循环不断的调用 write。这个调用,就是不停地做写的动作。这个和我们在上面看到的监控数据也是一致的。
六、总结
总结一下,通过这些源码说明,请你在使用的时候,要注意一下,像这样的工具,如果说只是为了单纯地消耗系统级的资源,然后观察应用在较少的可用资源下的表现如何,这样的工具是可以用的。
但是如果是想要模拟你的业务层出现的问题,那我劝你还是别用这样的工具了。
版权声明: 本文为 InfoQ 作者【zuozewei】的原创文章。
原文链接:【http://xie.infoq.cn/article/471e7cc64f24460313e851157】。
本文遵守【CC-BY 4.0】协议,转载请保留原文出处及本版权声明。
zuozewei
测试及性能领域创作者 2017.12.23 加入
「7DGroup」技术公众号作者,CSDN博客专家、测试领域优质创作者,华为云·云享专家,极客时间《全链路压测实战30讲》专栏作者之一,极客时间《性能测试实战30讲》、《高楼的性能工程实战课》专栏编委。
评论